Effects of Welding

on Health, IX

Key Words Welding, health, cancer, disease, exposure, fumes, gases, literature review, noise, radiation, toxicology

Effects of Welding
on Health IX

Research performed by Biomedical Toxicology Associates, Frederick, Maryland, under contract with the American
Welding Society and supported by industry contributions.
Performed by:
Winifred G. Palmer, Ph.D.
and
James C. Eaton, P.E.

Abstract
This literature review, with 232 citations, was prepared under contract to the American Welding Society for its Safety
and Health Committee. The review deals with studies of the fumes, gases radiation, and noise generated during various
arc welding processes. Section 1 summarizes recent studies of occupational exposures, Section 2 contains information
related to the human health effects, and Section 3 discusses the effects of welding on animals and cell cultures.

Prepared for
Safety and Health Committee
American Welding Society
550 N.W. LeJeune Road
Miami, Florida 33126

International Standard Book Number: 0-87171-462-0

American Welding Society, 550 N.W. LeJeune Road, Miami, FL 33126
1995 by American Welding Society. All rights reserved.
Printed in the United States of America.
This report is published as a service and convenience to the welding industry and is the product of an independent
contractor (Biomedical Toxicology Associates) which is solely responsible for its contents. The materials in this report
have not been independently reviewed or verified and are offered only as information. AWS assumes no responsibility
for any claims that may arise from the use of this information. Users should make independent investigations to
determine the applicability of this information for their purposes.

ii

Personnel
K. A. Lyttle, Chairman
J. F. Hinrichs, 1st Vice Chairman
M. T. Neu, MD, 2nd Vice Chairman
M. E. Kennebeck, Jr., Secretary
J. T. Ashe
W. J. Astleford*
K. L. Brown
O. J. Fisher*
S. S. Glickstein*
W. S. Howes
A. F. Manz
R. J. Simonton*
D. H. Sliney*
W. O. Thompson*
R. J. Tucker*
R. M. Tuggle*
M. J. Vasquez*

Praxair, Incorporated-Linde Division

A. O. Smith Corporate Technology
Caterpillar, Incorporated
American Welding Society
Snyder General Corporation
Southwest Research Institute
The Lincoln Electric Company
Consultant
Westinghouse Electric Corporation
National Electrical Manufacturers Association
A. F. Manz Associates
Consultant
U. S. Army Environmental Hygiene
U.S. EPA
Glendale Protective Tech
U.S. Department of Energy
Shell Oil Company

*Advisor

iii

Foreword
(This Foreword is not a part of Effects of Welding on Health IX, but is included for information purposes only.)
This literature review was prepared for the Safety and Health Committee of the American Welding Society to provide
an assessment of current information concerning the effects of welding on health, as well as to aid in the formulation and
design of research projects in this area, as part of an on going program sponsored by the Committee. Previous work
consists of the reports Effects of Welding on Health I through VIII each covering approximately 18 months to two years.
Conclusions based on this review and recommendations for further research are presented in the introductory portions
of the report. Referenced materials are available from:
Biomedical Toxicology Associates
P.O. Box 3539
Frederick, MD 21701
Tel. (301) 662-0783

iv

Comparative Listing Welding Processes

Explanatory Note: Terms used in the technical literature sometimes do not correspond to those recommended by
AWS in its publication ANSI/AWS A3.0, Standard Welding Terms and Definitions.
Accordingly, the following list may aid the reader in identifying the process in use.
EWH IX
Gas or Flame Cutting
Gas Welding
MAG
MIG, GMA
MMA, SMA
TIG
Wire

Preferred AWS Term

(OC)
(OFW)
(GMAW)
(GMAW)
(SMAW)
(GTAW)

Oxygen Cutting or (OFC) Oxyfuel Gas Cutting

Oxyfuel Gas Welding or (OAW) Oxyacetylene Welding
(with specified shielding gas)
Gas Metal Arc Welding
Shielded Metal Arc Welding
Gas Tungsten Arc Welding
Electrode

Acknowledgments
Funds for this project were provided by the American Welding Society.
The American Welding Society gratefully acknowledges the financial support of the program by industry contributions.

Supporting Organizations
Air Products and Chemicals, Incorporated
Airco Welding Products
Allis-Chalmers
Alloy Rods Division, The Chemetron Corporation
AWS Detroit Section
AWS New Orleans Section
Arcos Corporation
The Binkley Company
Caterpillar Tractor Company
Chicago Bridge and Iron Company
Grove Manufacturing Company, Division of Kidde, Incorporated
General Electric Company
The Heil Company
Hobart Brothers Company
INCO Alloys International
Lincoln Electric Company
Miller Electric manufacturing Company
National-Standard Company
A.O. Smith Corporation
Teledyne-McKay, Incorporated
Trinity Industries, Incorporated
Truck Trailer Manufacturers Association
Walker Stainless Equipment Company
Weld Tooling Corporation
Many other organizations have also made contributions to support the ongoing program from May 1979 to the present.

vi

Table of Contents
Page No.
Personnel ................................................................................................................................................................... iii
Foreword ................................................................................................................................................................... iv
Comparative Listing Welding Processes ............................................................................................................... v
Acknowledgments ....................................................................................................................................................... vi
Introduction ................................................................................................................................................................. 1
Executive Summary ..................................................................................................................................................... 3
Technical Summary ..................................................................................................................................................... 7
Conclusions ............................................................................................................................................................... 17

Section One The Exposure

1. Introduction ........................................................................................................................................................ 19
2. Fumes ................................................................................................................................................................. 19
2.1 Analytical Techniques ............................................................................................................................... 22
2.2 Analysis of Metals in Biological Tissues .................................................................................................. 22
3. Workplace Exposure Limits ............................................................................................................................... 22
4. Ozone .................................................................................................................................................................. 23
5. Electromagnetic Radiation ................................................................................................................................ 25
6. Electromagnetic Fields ...................................................................................................................................... 26
7. Incidental Exposures ......................................................................................................................................... 26
7.1 Production Coatings ................................................................................................................................. 26
7.2 Radiation Exposure ................................................................................................................................... 26
8. Hygiene and Work Practices ............................................................................................................................. 27
8.1 Ventilation and Air Cleaning Equipment ................................................................................................. 27
8.2 Protective Gear ......................................................................................................................................... 27
8.3 Accidents ................................................................................................................................................... 28
8.4 Stress ......................................................................................................................................................... 29

Section Two Effects of Welding on Human Health

9. Respiratory Tract ............................................................................................................................................... 29
9.1 Pulmonary Function and Bronchitis ......................................................................................................... 29
9.2 Case Reports ............................................................................................................................................. 32
10. Cancer ................................................................................................................................................................ 33
10.1 Lung Cancer .............................................................................................................................................. 33
10.2 Nasal and Laryngeal Cancer ..................................................................................................................... 37
10.3 Urogenital Tract Cancer ........................................................................................................................... 37
10.4 Cancer in Children of Welders ................................................................................................................. 37
10.5 Cancers Associated with Electromagnetic Fields .................................................................................... 38
10.6 Cancers Associated with Ultraviolet Radiation ....................................................................................... 38
11. Metal Fume Fever .............................................................................................................................................. 39
12. Effects on the Ear ............................................................................................................................................... 39
13. Effects on the Eye and Vision ............................................................................................................................ 40

vii

13.1
13.2
13.3
13.4
13.5

Cataracts .................................................................................................................................................... 40
Photokeratitis ............................................................................................................................................ 40
Eye Injury ................................................................................................................................................. 41
Long-Term Effects .................................................................................................................................... 41
Contact Lenses .......................................................................................................................................... 41

14. Effects on the Skin .............................................................................................................................................. 42

15. Effects on the Nervous System ........................................................................................................................... 42
16. Effects on the Immune System ........................................................................................................................... 43
17. Effects on the Musculoskeletal System .............................................................................................................. 43
18. Effects on the Kidney ......................................................................................................................................... 46
18.1 Cadmium .................................................................................................................................................... 46
18.2 Chromium .................................................................................................................................................. 46
19. Fertility ............................................................................................................................................................... 46
20. Teeth ................................................................................................................................................................... 49
21. Effects of Specific Metals ................................................................................................................................... 49
21.1 Beryllium .................................................................................................................................................. 49
21.2 Cadmium ................................................................................................................................................... 50
21.3 Iron ............................................................................................................................................................ 50
22. Biological Monitoring ....................................................................................................................................... 50
22.1 Chromium: Biomarkers ............................................................................................................................ 50
22.2 Chromium: Biological Monitoring ........................................................................................................... 51
22.3 Nickel ........................................................................................................................................................ 52
22.4 Aluminum ................................................................................................................................................. 52
22.5 Zinc ........................................................................................................................................................... 53
23. Incidental Exposures ......................................................................................................................................... 53
23.1 Degreasing Agents .................................................................................................................................... 53
23.2 Coated or Contaminated Surfaces ............................................................................................................ 53
23.3 Allergens ................................................................................................................................................... 54

Section Three Effects in Animals and Cell Cultures

25. Fertility ............................................................................................................................................................... 54
26. Fibrosis .............................................................................................................................................................. 55
27. In Vitro Tests ...................................................................................................................................................... 55
28. Effects of Welding Fumes in Animals ................................................................................................................ 56
29. Biochemical Studies ........................................................................................................................................... 56
References ................................................................................................................................................................. 59

viii

List of Tables
Table
1.
2.
3.
4.

Page No.
Lung Cancer in Welders, Cohort Studies ................................................................................................... 34
Lung Cancer in Welders, Case-Control Studies ........................................................................................ 35
Effect of Short-Term Exposure to Tri- and Hexavalent Chromium on Testicular Weight, and
Number of Spermatozoa per Epididymis in Adult Wistar Rats ................................................................ 55
Metal Content of Fumes from Electrodes Tested in Lipid Peroxidation Assays ...................................... 57

List of Figures
Figure
1A.
1B.
1C.
2.
3.
4.
5A.
5B.
5C.
6.

Page No.
Fume Generation Rate During GMAW of Mild Steel Using Solid Mild Steel Wire Showing
the Effects of Increasing Concentration of CO2 in the Shield Gas with Increasing Current .................... 21
Fume Generation Rate During GMAW of Stainless Steel Using Solid Stainless Steel Wire
Showing the Effects of Helium in the Shield Gas with Increasing Current .............................................. 21
FGR for GMAW of Aluminum Alloys Containing Either 5% Magnesium (Alloy No. 5636)
or 5% Silicon (Alloy No. 4043) Showing Effect of Magnesium in the Metal and Helium in the
Shield Gas with Increasing Current ........................................................................................................... 21
Cumulative Frequency Distribution of Welding Fume Concentrations in the Breathing Zone
of Welders During Welding without the Use of Local Exhaust Ventilation ............................................ 24
Frequency of Physical Symptoms Affecting Different Parts of the Musculoskeletal System ................. 44
Drawing of Hand Showing Location of Carpal Tunnel and Position of the Median Nerve
Affected by Carpal Tunnel Syndrome ....................................................................................................... 45
Cadmium Levels in Serum from Six Welders During the First 5 Years After Exposure to
Cadmium Ceased ........................................................................................................................................ 47
Cadmium Levels in Urine in Six Welders During the First 5 Years After Exposure to
Cadmium Ceased ........................................................................................................................................ 47
Concentrations of Beta 2-Microglobin in Urine of Six Welders During the First 5 Years
After Exposure to Cadmium Ceased .......................................................................................................... 48
Aluminum Concentrations in Urine from Two Welders During the First 4 Years After
Welding of Aluminum Ceased ................................................................................................................... 52

ix

Introduction
Protecting the health of workers in the welding environment is a major concern of the American Welding Society. To
stay abreast of this subject, the health literature is periodically reviewed and published in the report Effects of Welding
on Health. Eight volumes have been published to date; the first covered data published before 1978, while the latter
seven covered time periods between 1978 and December 1989. The current report includes information published between January 1990 and December, 1991. It should be read in conjunction with the previous volumes for a comprehensive treatment of the literature on the Effects of Welding on Health.
Included in this volume are studies of the characteristics of welding emissions that may have an impact on the control
technologies necessary to protect the welder (Section 1). In keeping with previous volumes, the health studies are organized according to the affected organ system. The respiratory tract, the primary route of exposure to welding emissions,
is also a major target organ of a number of components of these emissions. Acute (e.g., metal fume fever, cadmium poisoning) as well as potential chronic respiratory effects (e.g., bronchitis, cancer) of welding emissions are of concern.
Chronic effects on other systems such as the urogenital tract have also been studied. One such effect, injury of the kidney
tubules, is known to result from chronic exposure to cadmium. The effect of welding on fertility has received recent attention, but the evidence for a negative impact of welding on fertility is weak and, at the most, uncertain. Continued research in the form of epidemiologic studies, investigations with laboratory animals, and in vitro cell studies will help to
resolve these questions.

Executive Summary

The Respiratory Tract

welders included asthma and dyspnea. Chinn et al. (Ref.

48) and Bogadi-Sare (Ref. 31) found that breathlessness
on exertion was significantly more frequent among welders. Welding was also found to be significantly associated with asthma (Ref. 11).

The contribution of welding to the development of

pulmonary function deficits and respiratory disorders
such as bronchitis and asthma remains uncertain, and
conflicting results have been reported by different investigators. In studies which have identified a positive relationship between welding and respiratory disorders, it
has not been possible to identify specific components of
the exposure responsible for these conditions.
Chinn et al. (Ref. 48) and Bogadi-Sare (Ref. 31)
found a statistically significant decrease in ventilatory
parameters indicative of bronchial obstruction in welders. These results were in accord with those of Kilburn et
al. (Refs. 111, 112) who found that reductions in lung
function in welders were small, but statistically significant. Lukac et al. (Ref. 128) also found decreases in
FEV1 and FEV1/FVC among welders, especially among
those who smoked. However, non-welding controls were
not examined in that study.
In contrast to these results, Melbostad and Ruud (Ref.
136), in their study of metal and machine workers, and
Demers et al. (Ref. 53), in their study of boilermakers,
found no association between welding and deficits in
FVC or FEV1. Rossignol et al. (Ref. 176) actually found
an increase in FEV1/FVC in experienced welders.
Inconsistent results also were seen among studies of
respiratory symptoms. While chronic bronchitis was
shown to be significantly related to smoking (Ref. 48),
Chinn et al. (Ref. 48) and Bogadi-Sare (Ref. 31) found
that it was not related to trade as a welder. However,
Kilburn and Warshaw (Refs. 111 and 112) found that
nearly 20% of electric arc welders had chronic bronchitis, and the incidence of this disease was related to welding, especially in nonsmokers. In other studies, welders
were found to have a high incidence of chronic bronchitis (Refs. 126 and 128). Kleiner et al. (Ref. 115) found
that welders developed chronic bronchitis at an earlier
age and after briefer employment than mechanics. This
effect was much greater in smokers than in non-smokers.
Other respiratory symptoms that were studied among

Cancer
The International Agency for Research on Cancer
(IARC) evaluated epidemiologic studies reported
through 1989 which assessed the incidence of cancer in
welders (Ref. 100) and concluded that there is limited
evidence in humans for the carcinogenicity of welding
fumes and gases. This evaluation indicates that a positive association has been observed between exposure to
[welding fumes] and cancer for which a causal interpretation is considered...to be credible, but chance, bias, or
confounding could not be ruled out with reasonable confidence. The evidence for carcinogenicity in experimental animals was judged to be inadequate and the overall
evaluation of IARC was that welding fumes are possibly carcinogenic to humans.
Eleven cohort and twelve case-control studies were
considered by IARC in their evaluation of the cancer risk
of welders. While most of the cohort studies showed a
greater incidence of lung cancer among welders than
among control populations, only one of these was significant (Ref. 186). Six of the case-control studies examined
by IARC showed greater than a 100% excess lung risk.
The excess risk was statistically significant in four of the
latter studies. These four studies, served in part as the
basis for the conclusion by IARC that there is limited evidence for the carcinogenicity of welding emissions in
humans. Also considered by IARC was the positive association of lung cancer with exposure to welding fumes
that was seen in an analysis of data pooled from 21 casecontrol and 27 cohort studies conducted between 1985
and 1989. The results of that analysis were published in
1991 by Simonato et al. (Ref. 184). In this large multinational investigation, the combined study population con3

sisted of 11,092 welders from 135 companies in 9

countries. Of the major causes of death, only those from
malignant neoplasms were higher than expected, based
on comparison with control populations. This excess was
of borderline significance, and was due primarily to a
significant excess of deaths from lung cancer (SMR =
134). The risk of death from lung cancer was higher in
mild steel only welders (SMR = 178) than in stainless
steel ever welders (SMR = 128). A statistically significant excess of bladder cancer was also observed (SMR =
191). The lung cancer incidence tended to increase with
time since first exposure for the predominantly stainless
steel group. This trend was not significant for any other
group of welders. Estimated cumulative doses of total
fume, total chromium, or hexavalent chromium were not
significantly associated with mortality from lung cancer.
The conclusions of this study were confounded by the
presence of five cases of mesothelioma, indicative of asbestos exposure, in the cohort of welders. In addition, the
effects of tobacco smoking could not be ruled out.
Four new epidemiologic studies of the association between cancer and welders were reported in 1991. Two of
these found no cancer excesses compared with the general population (Refs. 25 and 192) and the other two
studies showed an excess lung cancer incidence among
welders (Refs. 69 and 178).

Fertility
Bonde conducted a series of studies concerned with
the fertility of welders. In a case-control study, he
showed an increased risk for subfertility (delayed conception) among 432 male welders compared with 240
nonwelding metal workers and electricians (Ref. 32). In
a cohort study, Bonde et al. (Ref. 35) showed that among
persons who had ever worked as welders, the probability
of having a child was slightly, but significantly, reduced
during the year following a year of welding exposure.
The reduction in fertility was associated with the welding
of mild steel, but not with SMAW or GTAW of stainless
steel. Changes in sperm quality and follicle stimulating
hormone were found in 35 stainless steel welders and
also in 46 mild steel welders compared with 54 nonwelding metalworkers (Ref. 33). The effects were most
marked among mild steel welders. Changes in semen
quality were not found to be reversible, as semen quality
did not improve during a 3-week welding-free vacation
period (Ref. 34).
While Bondes studies suggest a reduction in fertility
and semen quality with welding, the changes observed
were slight, albeit significant. A positive dose-response
was not obtained when fertility was considered in terms
of years of welding exposure. Nonwelders were ex-

cluded from the large cohort study (Ref. 35) because

their fertility rate was lower than that of welders during
periods of nonexposure. The rationale for excluding nonwelders from the study is not clear, and it is possible that
a direct comparison between the overall fertility rates
among welders (including periods of both welding and
nonwelding) and nonwelders would have shown no difference between the two groups.
Bondes conclusion that welding reduces fertility was
based on differences in birthrates between periods of
welding and nonwelding in the same group of welders.
However, differences in the fertility of welders between
periods of welding and nonwelding could have been due
to factors other than welding, and a more valid comparison would be between men who were never welders and
those who worked continuously as welders during their
entire family-raising period. Finally, while the investigation of semen quality supported the concept that welding
is associated with decreased fertility, there were some inconsistencies between the findings of the fertility and
semen quality studies. As Bonde pointed out, the observed changes were small and thus subject to confounding factors which are impossible to control with human
populations. Because the effects are small, controlled
studies in laboratory animals may be necessary to resolve
questions concerning the effects of welding exposures on
fertility.
One animal study published during this report period
examined the effects of chromium on male genitalia
(Ref. 63). In that study, hexavalent chromium, but not
trivalent chromium caused a reduction in testicular
weight, a dose-dependent increase in the number of atrophic seminiferous tubules, and a reduction in the epididymal sperm count. This study is not directly relevant to
fertility in welders because the route of exposure was intraperitoneal injection, the effects of which can vary
markedly from those caused by inhalation.

Chromium
An important area of research involves the effects of
long-term exposure to low levels of chromium. This pursuit has captured the attention of investigators concerned
with industrial exposures such as welding (Ref. 36 and
219) as well as environmental exposures (Ref. 41).
Drawing from the known effects of acute exposure to
high concentrations of chromium on the kidney, Wedeen
and Qian (Ref. 219) conjectured that long-term exposure
to low levels of chromium could cause kidney tubular
disease. Based on the fertility studies discussed above,
Bonde postulated that long-term exposure to low levels
of chromium could reduce fertility in welders. In regard
to this issue, Bonde and Christensen (Ref. 36) deter-

mined chromium levels in body fluids from welders

working with processes that generate fumes containing
low concentrations of chromate. Chromium concentrations were significantly higher in blood and urine from
stainless steel and mild steel welders than from controls.
However, the concentrations of chromium in blood and
urine did not change across a workshift or after a 3-week
vacation break from welding. The authors found these
data to be consistent with a gradual buildup of chromium
in the body during long-term welding exposure. While
high concentrations of chromium were found in seminal
fluid from welders conducting SMAW of stainless steel,
the extremely wide variation in chromium levels indicated that much of the chromium may have been derived
from nonoccupational activities.
Because studies in which effects are low and difficult
to detect are easily confounded by external factors unrelated to occupational exposures, it is important to be
aware of nonvocational factors that could introduce error
into epidemiologic studies. As part of their studies of biological monitoring of New Jersey state workers with
low-level intermittent exposures to chromium in soil,
Bukowski et al. (Ref. 41) determined the contribution of

avocational activities and personal habits to urinary chromium levels. They showed that males had slightly higher
urinary chromium levels than did females. Beer drinking
was associated with a significant increase in urinary
chromium. Subjects who used tobacco, and those who
had exercised within 24 hours before sampling, had
slightly lower chromium levels in the urine. Neither
welding nor engaging in hobby activities with possible
chromium exposures was associated with increased concentrations of chromium in the urine or blood.
Coogan et al. (Ref. 50) demonstrated that lymphocytes can serve as a better biomarker than red blood cells
for long-term chromium exposure. They showed that
white blood cells can accumulate significantly more
chromium than red blood cells. While the chromium uptake by red blood cells appeared to be independent of the
valence state of the administered chromium, lymphocytes accumulated only hexavalent chromium. The investigators concluded that the exclusive accumulation of
hexavalent chromium by white blood cells supports their
use as target cells in the development of biomarkers for
assessing exposure to chromium.

Technical Summary

The Exposure

GMAW of steel coated with tin or zinc produced more

fumes than welding aluminum-coated or uncoated steel
(Ref. 135). Medack and Heinze (Ref. 135) calculated
that air concentrations of zinc released when welding
zinc-coated steel, would approach the PEL before any of
the other components of the fume. Lead would be the
comparable limiting factor from welding tin-coated steel.
In fumes from welding aluminum-coated or uncoated
steel, manganese from the electrode was the limiting
factor.

Fumes
Using a model developed to predict the composition
of welding fumes generated by SMAW and FCAW,
Hewitt and Hirst (Ref. 87) showed that the flux contributes disproportionately to the mass of the fume based
upon its share of the total weight of the wire. Hilton and
Plumridge (Ref. 91) demonstrated that gas metal arc
welding (GMAW) of stainless steel with a carbon dioxide shielding gas or stainless steel with helium as the
shielding gas produced much more fumes than did
GMAW with argon-based shielding gases.
When welding aluminum alloys at currents above
150 A, the FGR from an alloy containing 5% magnesium
was much greater than that from a 5% silicon alloy.
Chromium, nickel, manganese, and fluoride were the
components most often measured in several studies of
the hazardous components of welding fumes. The hazardous fume component that was most likely to reach the
permissible exposure limits (PEL) during welding of
stainless steel was hexavalent chromium [Cr(VI)] (Refs.
152, 194, 213, and 230).
Sutton (Ref. 194) showed that Cr(VI) produced by
GMAW and gas tungsten arc welding (GTAW) of stainless steel was much lower than that from shielded metal
arc welding (SMAW) of the same material. However,
Olah and Pospisilova (Ref. 152) found that even though
total fume production from GMAW is less than that from
SMAW using electrodes with a high-chromium content,
the chromium concentration in the GMAW fumes was so
much higher that chromium emissions were about equal
for the two processes. Although GTAW fumes contained
up to 21% chromium, GTAW produced the lowest emission rate for chromium of all the methods tested. Zaks
(Ref. 229) showed that manganese (from acid coatings)
and fluoride (from basic coatings) were the fume components most likely to reach hazardous levels from SMAW
of mild steel.

Analytical Techniques. A particle-sizing method used

by Ukkonen et al. (Ref. 206), combining a photometer
and an electric particle counter, worked well for sizing
particles with unit density but not for the denser particles
of welding fumes. Battistoni et al. (Ref. 17) used a photometer to measure particle concentrations from GMAW
with a continuous automatic welder. Goschnick et al.
(Ref. 80) used secondary neutral mass spectrometry to
measure the compositions of successive layers of welding particles. Chromium was enriched in the fume particles compared to its concentration in the electrodes.
Contrary to findings of other investigators (Refs. 81 and
129), the chemical composition of the outer and inner
layers of the particles did not vary.

Lasers
Lasers used for welding are usually class IV, the most
hazardous rating, and require safety features to prevent
exposure. Ordinary welding curtains are not suitable for
protection against the energy of the laser beam, and special laser barriers must be used. A useful feature that can
be incorporated into these barriers is a surface coating
that shows visual signs of exposure to stray laser beams
(Ref. 221). The electrical power supplies and the chemicals used with lasers also present hazards in the workplace, as do the fumes generated by welding with lasers
(Ref. 109). Engel et al. (Ref. 60) found that more fumes
7

were released while cutting stainless steel with CO 2

lasers than while cutting nonalloyed steel. The fume particles were spheres and agglomerates of spheres of submicrometer size, and their composition was similar to
that of the metal being cut.

Electromagnetic Fields
Electromagnetic fields (EMFs) are produced by the
power sources used in all types of electric arc welding.
Although it is not clear that EMFs from welding pose a
health hazard, it has been suggested by Zyubanova et al.
(Ref. 232) that their intensity may be diminished by the
use of coaxial cable to minimize the distance between the
electric power lines.

Incidental Exposures
Henriks-Eckerman et al. (Ref. 84) developed a sampling procedure for monitoring workplace pollutants
arising from production coatings on the metal surface.
The procedure, which included a chemosorbent tube for
aldehydes and a tube of adsorbent resin preceded by a
glass fiber filter for particulates and condensable organic
chemicals, was used to monitor the exposure of metal
workers to paint degradation products generated by
welding, flame cutting, and straightening painted steels
in a Finnish shipyard (Ref. 61).
Onodera et al. (Ref. 157) measured emissions of radioactivity during the decommissioning of a Japanese
nuclear power plant. Generally, cutting with a welding
torch produced less radioactivity than cutting with mechanical tools. Cutting under water reduced radioactive
emissions to an even greater extent. Carmichael and
Haynes (Ref. 46) described safety practices and exposure
measurements during repairs to apparatus contaminated
with tritium at a Canadian nuclear power plant.

Hygiene and Work Practices

Equipment. Using computer simulations and laboratory
data, Tum Suden et al. (Ref. 205) showed that the parameters of hood design and operation that affected breathing zone concentration are flow rate through the hood,
hood aspect ratio, and the welders position relative to
the hood. Jakubcik (Ref. 102) reported that ventilation
was improved during tack-welding of long pieces by enclosing the work in a segmented hood which could be
opened in the area where the work was being performed.
For gun-mounted fume collection devices, the position
of the exhaust vents is important in determining collection efficiency (Ref. 51). A welding table designed by

van der Veen and Regensburg (Ref. 210) combined efficient ventilation, absorption of ultraviolet (UV) radiation, and height adjustments to prevent muscular stress.
Protective Gear. Two devices employing activated carbon filtration were shown to be effective in reducing
ozone concentrations in the breathing zone (Ref. 193).
Several types of material were found to provide superior
heat protection to asbestos, but some of the flame-proof
clothing tested lost resistance to ignition after 10 or 20
washings (Ref. 105).
Accidents. Analysis of accidents fatal to welders
showed that the welding apparatus was not often implicated. Most of the fatal accidents analyzed by Trent and
Wyant (Ref. 203) were due to welders falling or being
struck by falling objects and to environmental hazards in
the workplace. Three fatal accidents involving falls were
described by the National Institute of Occupational
Safety and Health in 1990 and 1991 (Refs. 146-148).
Rekus (Ref. 169) reviewed several cases of fatal accidents to workers entering confined spaces. Tanks and
holds that have been recently opened after having been
sealed for long periods are particularly hazardous due to
accumulation of toxic gases or depletion of oxygen.
Stress. Physiological stresses to welders were documented in two studies in 1990 and 1991. Valente and
Chiapperini (Ref. 209) and Richter (Ref. 170) measured
pulse rate, blood pressure, oral and skin temperature, and
perspiration rate in four welders and related the results to
working conditions. Psychological stresses among welders were attributed by Richter (Ref. 170) and Hyytiainen
and Uutela (Ref. 97) to the tediousness of the work, the
constant need for exercising caution, and the isolation of
the worker in the welding environment. Gerhardsson
(Ref. 75) found that welders reported a combination of
high work load and little opportunity to influence the
work situation, which has been related to some physical
manifestations of stress.

The Effects of Welding

on Human Health
Respiratory Tract
Pulmonary Function and Bronchitis. Welders were
included in seven studies of lung function and respiratory
symptoms. Chinn et al. (Ref. 48) found significant associations between trade as a welder and development of
breathlessness upon minimal exertion. A decline in the
lung function parameter FEV1 was related to occupation

as a welder or caulker/burner and to atopy (positive response to skin tests with common allergens).
Bogadi-Sare (Ref. 31) found a significant decrease in
FEV1 and FEV1/FVC among 47 stainless steel welders
and 59 nonwelders employed in machining and polishing
stainless steel. The frequency of chronic bronchitis was
the same in exposed and unexposed workers, but the exposed group had significantly more frequent complaints
of dyspnea and choking. In a community study which included 84 welders, Bakke et al. (Ref. 10) found welding
to be significantly related to chronic obstructive lung disease and asthma.
Kilburn and Warshaw (Ref. 111) found small, but significant deficits in FEV1, FEF25-75, and FEF75-85 in 291
male welders compared with historical controls from the
same geographical area. In a second study, Kilburn et al.
(Ref. 112) found that the rate of chronic bronchitis was
substantially elevated in gas tungsten arc welders and
was unrelated to smoking. There was a small but significant reduction in FVC, FEV1, and FEF25-75 in smoking
welders compared to the reference population while only
the FEF25-75 was reduced in nonsmoking welders. Further tests showed that changes in lung function do not
occur during a workshift in workers exposed to stainless
steel welding fumes.
Lukac et al. (Ref. 128) found deficits in FEV1 and
FEV1/FVC that were related to duration of exposure in
welders of steel bridge components. This study did not
have nonwelding controls. Melbostad and Ruud (Ref.
136) found no difference in lung function (FEV1/FVC)
between welders and machinists, but work-related respiratory symptoms were more prominent among welders.
Contrary to the findings of most other studies, Rossignol
et al. (Ref. 176) found a significant increase in
FEV1 /FVC with length of employment as a welder or
burner.
Kleiner et al. (Ref. 115) found significantly more
chronic bronchitis among welders (24.3%) than among
mechanics. Demers et al. (Ref. 53) studied asbestos-exposed boilermakers and did not find a positive relation
between welding and respiratory disease or changes in
lung function. Lubianova et al. (Ref. 127) found a 35%
incidence of signs of chronic bronchitis among welders,
related to years of exposure, age, and smoking.

tis, and massive emphysema in the lungs of a man who

had worked as an arc welder for 15 years. Chemical analysis of his lung tissue revealed excessive concentrations
of iron, aluminum, chromium, nickel, manganese, and
copper. Pneumoconiosis and asbestosis lesions in the
lungs of a welder examined by Kishimoto et al. (Ref.
113) were considered to provide an ideal culture medium
for a Mycobacterial infection which was present in the
patients lung.

Case Reports. Three of the 35 cases of occupational

asthma reported to the Ministry of Labor in Singapore
between 1983 and 1990 were welders. The prevalence of
smoking and the percentage who reacted positively to
environmental allergen skin prick testing were not significantly different from controls (Ref. 121).
Lasfargues et al. (Ref. 120) described a case of siderosis and interstitial fibrosis in a man who had welded for
30 years. Wagner et al. (Ref. 217) described post mortem
findings of pneumonoconioses, chronic tracheobronchi-

Cancer at Other Sites. Ahrens et al. (Ref. 2) found that

nasal and laryngeal cancer were strongly linked to smoking and alcohol consumption but not to welding. Yu et al.
(Ref. 227) identified exposure to combustion products
among risk factors for nasopharyngeal carcinomas. In
this exposure group, 15 cases and eight controls were exposed to welding fumes, but the significance of these
numbers was not reported. Bladder cancer was found to
be significantly elevated in the combined study population analyzed by Simonato et al. (Ref. 184), but the

Cancer
Lung Cancer. The International Agency for Research
on Cancer (IARC) concluded that welding fumes are
possibly carcinogenic to humans, based upon limited evidence in humans and inadequate evidence in experimental animals (Ref. 100). Included in the data considered by
IARC was the analysis of a pool of 21 case-control studies and 27 cohort studies published in 1991 by Simonato
et al. (Ref. 184). That analysis showed a significant excess of lung cancers (Standard mortality ratio: SMR =
134) among welders.
Four new epidemiologic studies of the association between cancer and welders were reported in 1991. Becker
et al. (Ref. 25) found a significant excess of deaths from
all malignancies among welders when compared with
machinists in the same plant in the Federal Republic of
Germany (Relative risk = 1.6). However, welders did not
have a significantly elevated risk of lung cancer when
compared to the general population. In an American
study (Ref. 192), limited to mild steel welders who had
no asbestos exposure, lung cancer was not associated
with welding when welders were compared with either
the general population or to nonwelders in the same
plants. Analysis of cancer registry data (1971 to 1980) in
Finland found a statistically significant increased risk for
lung cancer among welders (Standardized Incidence
Ratio = 150) (Ref. 178). Similarly, the 125 lung cancer
deaths of welders in British Columbia between 1950 and
1984 were significantly greater than expected (Proportional Mortality Ratio = 129), compared with the general
population and to a subgroup of skilled manual workers.
Significantly greater risks were also found for Hodgkins
disease and all cancers in that study (Ref. 69).

10

incidence of cancer was unrelated to duration or intensity

of exposure.
Cancer in Children of Welders. A heritable form of retinoblastoma, a tumor in the retina occurring in young
children, was linked to fathers occupation in the metalworking industry. The nonheritable form of the disease,
believed to be the result of post-conception exposures,
was linked to fathers working in a job cluster that included welders, machinists, and paper-processing workers (Ref. 43). Welders were not analyzed separately in
either study so that the contribution of welding exposures
to this disease cannot be assessed.
Cancers Associated with Electromagnetic Fields. Five
epidemiological studies of the incidence of leukemias
and cancer of the central nervous system reported in
1990 and 1991 included welders in the exposed population. Leukemia was linked to exposure to electromagnetic fields (EMFs), but not to welding, in three studies,
(Refs. 16, 106 and 201) and no relationship was found
between the risk of leukemia and exposure to EMFs in
the second study (Ref. 158). EMF exposure was implicated in central nervous system cancer in three studies
(Refs. 68, 106 and 201). Although the data for welders
were analyzed separately in all three studies, only one indicated a link between welding and central nervous system cancer, and that was of borderline significance (Ref.
201).
Cancers Associated with Ultraviolet Radiation. Holly
et al. (Ref. 93) conducted a case-control study of the association between uveal melanoma and exposure to UV
light. A statistically significant relative risk for uveal
melanoma was found among persons included in a group
who had experienced welding burn, sunburn of the eye,
or snow blindness.

Metal Fume Fever

A typical case of metal fume fever was described by
Heydon and Kagan (Ref. 88). The patient developed a
cough, chills, cramps, and difficulty breathing 2 hours
after cutting galvanized steel with a gas torch without
using a mask.
Based on the hypothesis that polymorphonuclear leukocytes (PMNs) responding to inhaled zinc oxide fumes
release mediators (cytokines) that can elicit the systemic
effects characteristic of metal fume fever, Blanc et al.
(Ref. 27) analyzed the white blood cell population and
concentration of two cytokines (interleukin-1 and tumor
necrosis factor (TNF) in bronchoalveolar lavage (BAL)
fluid following exposure of experienced welders to
fumes generated by welding galvanized mild steel.
While the population of inflammatory cells increased in

the BAL fluid during the first 22 hours after welding, little or no TNF or interleukin-1 was detected. The investigators concluded that they had examined the wrong
cytokines and postulated that other cytokines, or a
cytokine-like mechanism, may mediate the syndrome of
metal fume fever.

Effects on the Ear

Thirty cases of burns to the ears inflicted by flying
sparks from welding operations were discussed by two
investigators (Refs. 64 and 137). Fisher and Gardiner
(Ref. 64) ascribed these injuries to the lack of ear protection afforded by visor type helmets when working
overhead or in cramped conditions with the neck flexed.

Effects on the Eye and Vision

Using a theoretical model of the human eye, Okuno
(Ref. 151) concluded that infrared (IR) cataracts result
from heat conduction to the lens following absorption of
IR radiation by the cornea. The threshold IR irradiances
for cataract formation were determined to be in the range
of 163 to 178 mW/cm2 for exposures lasting longer than
5 minutes under normal conditions. To protect against cataract formation, Okuno recommended that IR irradiance
be limited to 80 mW/cm2 for exposures over 5 minutes.
Four cases of welders with cataracts were described
by Hanke and Karsten (Ref. 83). The IR irradiance received by the eyes of each of the welders was estimated
from detailed occupational histories and reconstruction
of their workplace routines. All four welders had routinely been exposed to doses of IR radiation in excess of
the maximum safe limits recommended by Okuno (Ref.
151).
An unusual case in which a welder developed bilateral maculopathy (retinal injury) in the absence of photokeratitis, following unprotected exposure of less than
two minutes duration to a SMAW unit was described by
Power et al. (Ref. 166). The investigators suggested a
link between the severity of the retinal lesions in the absence of keratitis with the drug fluphenazine, which the
patient had taken for 10 years for treatment of depression. They postulated that fluphenazine accumulates in
the retinal pigment epithelium, where it acts as a photosensitizing agent, making the retina particularly susceptible to ph otochem ical damag e. Neki (Ref. 144 )
questioned the assumptions made by Power et al. and
suggested that, in the absence of experimental proof, it is
just as likely that the fluphenazine accumulates in the
cornea where it provides protection against the action of
electromagnetic radiation.

11

Surveys of persons appearing in hospitals with eye injuries showed that 17% of the patients treated for eye injuries in a Hospital in Denmark were welders (Ref. 9)
and 11.5% of those treated for ocular foreign bodies during a six month period in a hospital in England were
welders (Ref. 13). A survey of welders conducted by Ten
Kate and Collins (Ref. 198) related eye injury to the use
of appropriate eye protection. Forty percent of the welders reported receiving at last one welding flash. Reports of eye symptoms (e.g., tired, sore, watery or itchy
eyes) correlated well (p<.001) with the number of flashes
received. Unprotected exposure to nearby welding was
experienced by 64% of the welders.
Narda et al. (Ref. 142) found that the frequency of
chronic conjunctivitis was substantially higher among
welders than controls who worked in a foundry in Italy.
The incidence of conjunctivitis was higher in welders
who used SMAW than in those using submerged arc
welding (32.4%). Norn and Franck (Ref. 149) found that
spheroid degeneration (small globular, yellowish lesions
on the exposed part of the conjunctiva or cornea), was
more common in welders than in other workers. A significant, but much smaller increase in the incidence of
pinguecula (yellow spots on the exposed conjunctival
bulbi) was also noted.

Wechsler et al. (Ref. 218) conducted a pilot study to

evaluate a range of occupational and environmental
agents for potential associations with Parkinsons disease. The study compared 34 Parkinsons disease patients (average age 68.4 years) with 22 neurology clinic
patients (average age 58.9 years) having diseases other
than Parkinsons disease. Analysis by metal exposure in
males indicated more frequent exposures to aluminum
and copper. Three of the male Parkinsons disease patients and none of the controls had been employed as
welders. The three welders reported exposures to aluminum but not to other metals.
Armon et al. (Ref. 7) conducted a case-control study
to evaluate risk factors for amyotrophic lateral sclerosis
(ALS). Men with ALS had spent significantly more time
welding or soldering than controls (p<0.01).

Effects on the Immune System

Ulrich et al. (Ref. 207) found significant changes in
humoral immunity in welders; immunoglobulin IgG increased, and IgM decreased with the duration of welding.
The serum indicators of inflammation, complement and
alpha-1-antitrypsin, changed significantly with the number of years welding experience.

Effects on the Skin

Two cases were described in which welders reacted to
reflected UV radiation. The first case involved a welder
who developed recurrent mild facial erythema. His condition was attributed to UV radiation reflected from a
white textile hood that he wore in addition to a helmet
(Ref. 92). A second welder developed facial dermatitis
associated with welding. Skin testing showed an abnormal reaction to UV radiation. A shield was used while
welding, so his dermatitis was ascribed to indirect exposure to UV radiation (Ref. 226).

Effects on the Nervous System

Sjogren et al. (Ref. 189) conducted a study of neuropsychiatric symptoms among 65 welders who welded primarily aluminum and 217 railroad track welders.
Subjects who welded aluminum, lead, or manganese for
long periods experienced significantly more neuropsychiatric symptoms than welders exposed to chromium or
nickel. Depression and difficulty concentrating were associated with exposure to aluminum fume. Short memory, forgetfulness, and frequent headache were
associated with welding lead-painted steel. Experiencing
painful tingling sensations was associated with exposure
to manganese fume.

Effects on the Musculoskeletal System

Torner et al. (Ref. 200) found that the incidence of
neck and/or shoulder symptoms was more prevalent
among welders than clerks from the same plant. The
range of motion in external rotation of the shoulders was
significantly less among the welders than the clerks and
almost half (47%) the welders had atrophied shoulder
muscles. Most of the welders (60%) considered their
symptoms to have been caused by heavy work for a prolonged time, while only 12% attributed their musculoskeletal symptoms to a sudden incident at work.
Prolonged or awkward working postures and incorrect heavy lifts and stumbling or falling were found to be
the major risk factors for accidents involving lower back
injuries in welders (Ref. 96). In addition, certain stress
factors, (i.e., hurry, monotony, and an accelerated working pace) were also related to lower back pain (Ref. 97).
Occupationally related lower back pain was infrequently
related to sudden movements.
Jarvholm et al. (Ref. 103) used electromyography and
simultaneous measurement of intramuscular pressure to
study the effectiveness of arm support in reducing the
load on the supraspinatus muscle of the shoulder during
simulated welding operations. It was determined that
arm support greater than 2.2 to 3.4 pounds (force) would

12

be needed to reduce intramuscular pressure sufficiently

to lower the incidence of shoulder pain and impairment.
Marciniak and Badowski (Ref. 132) compared spinal
X-rays among job applicants and experienced manual
workers in a Polish automobile factory and found that the
incidence of scoliosis was highest among workers in jobs
that imposed the greatest spinal loading, i.e., tinsmiths,
pressers, and welders.
Carpal tunnel syndrome (CTS) can result from repetitive wrist movement tasks such as welding (Ref. 160) or
from use of vibrating hand-held tools (Ref. 82). Hagberg
et al. (Ref. 82) found that welding was one of the four
most common occupations among patients with carpal
tunnel syndrome. Panio (Ref. 160) described a workplace program for the prevention of CTS that included
pre-employment screening; worker training; modification of job routine to reduce unnatural angles of the
wrist; modification of tools; and gloves or braces to
maintain correct wrist alignment.

Effect on the Kidneys

Cadmium. Trevisan and Maso (Ref. 204) examined the
reversibility of proteinuria in six cadmium-exposed welders. All of the subjects had worked in shops where brazing was performed with high-cadmium alloys and had
participated in an earlier study of urine proteins before
they ceased welding in 1982. Three of the subjects had no
indications of low molecular weight proteinuria before
they had ceased welding and their urinary protein levels
remained low during the 5-year follow-up period. The remaining three welders had shown low molecular weight
proteinuria, indicative of kidney tubular dysfunction,
while still exposed to cadmium. During the follow-up
period, beta 2-microglobinuria increased in two of these
welders, while in the third there was a temporary increase
in beta 2-microglobinuria which returned to normal levels
after 5 years of nonwelding. These results indicate that
cadmium-induced tubular dysfunction is reversible in
some, but not all, persons with this condition.
Chromium. While acute kidney disease can result from
massive exposure to chromium, chronic renal disease resulting from occupational or environmental exposure to
chromium has not been reported. Wedeen and Qian (Ref.
219) maintained that the effects of chronic chromium exposure on the kidneys have not been adequately studied.
He stressed the need for large-scale, prospective casecontrol epidemiologic studies to demonstrate whether or
not delayed renal effects can result from low-level, longterm exposure to chromium.

Fertility
Boshnakova and Karev (Ref. 37) found that spontaneous abortions occurred significantly more frequently in
the families of welders and that still births occurred significantly more frequently in families of nonwelding
manual control workers.
Bonde (Ref. 32) observed an increased risk for subfertility (delayed conception) in a case-control study of 432
male welders and 240 nonwelding metal workers and
electricians. In a continuation of that study, Bonde et al.
(Ref. 35) examined fertility in a Danish cohort of 3702
male metalworkers. Among subjects who had ever
welded, the probability of fathering a child during years
when they were not welding was significantly greater
than that of metalworkers who had never welded.
Among persons who had ever worked as welders, the
probability of having a child was slightly, but significantly, reduced during the year following a year of welding exposure. The reduction in fertility was associated
with the welding of mild steel but not with SMAW or
GTAW of stainless steel.
Bonde (Ref. 33) compared semen quality and serum
levels of sex hormones among 35 stainless steel welders,
46 mild steel welders and 54 non-welding metalworkers.
Changes in sperm quality and follicle stimulating hormone were seen in both groups of welders. Effects were
most marked among mild steel welders. To determine if
the changes in semen quality were reversible, semen
quality was examined in 19 of the mild steel welders, 18
of the stainless steel welders and 16 of the non-welding
metalworkers before and after a 3-week, welding-free
vacation period (Ref. 34). No significant improvements
in any of the semen parameters were observed at any of
the post-vacation measurements relative to the pre-vacation values.

Effects on the Teeth

Tatintsyan and Abgaryan (Ref. 197) reported that
87% of welders have some degree of periodontal disease
and gingivitis. They correlated this with a lower than
normal concentration of lysozyme, an enzyme destructive to the cell wall of some bacteria, in saliva collected
from the welders. These investigators developed a small
filtration device to protect the mouth from exposure to
welding fumes. After using the device for 2 years,
lysozyme increased in saliva collected from welders.

Effects of Specific Metals

Beryllium. A typical case of chronic beryllium disease
was described by Monie and Roberts (Ref. 138) in a

13

fifty-one-year-old Scottish welder who had a long-standing cough with morning production of mucoid phlegm.
His chest X-ray showed reticulo-nodular shadowing. Examination indicated reduced pulmonary function and biopsies showed mild fibrosis and epithelioid cell
granulomas. Beryllium was identified in biopsied lung
tissue.
Cadmium. Yates and Goldman (Ref. 225) described a
case of acute cadmium poisoning which occurred following two weeks of brazing ship propellers with an oxyacetylene torch using an alloy containing 20% cadmium.
After 10 days, the patient began to experience malaise
and breathlessness, developed fever and joint pain, and
had reduced lung function.
Fuortes et al. (Ref. 67) described an acute fatality in a
man who was using a propane torch and soldering gun to
join sheet metal surfaces. He died 3 days after developing an extremely high fever, cough, and abdominal pain.
Autopsy revealed pulmonary edema and congestion. Exceedingly high cadmium levels (280 ng/ml) were detected in his blood and milligram quantities of cadmium
were identified on the soldering gun tip and sheet metal
samples with which he had been working.
Iron. Lubianova (Ref. 126) found that the concentrations of iron and transferrin were elevated in the blood of
welders. Welders without pneumoconiosis had less blood
iron than welders with pneumoconiosis.

Biological Monitoring
Chromium: Biomarkers. Popp et al. (Ref. 165) examined DNA integrity in blood lymphocytes obtained from
39 stainless steel welders and 18 controls. The average
sister chromatid exchange (SCE) frequency was significantly lower for welders than controls. However, alkaline
filter elution indicated the presence of DNA-protein
cross-links which showed that SCE analysis alone could
lead to the false impression that DNA integrity remains
intact in chromium-exposed workers.
Coogan et al. (Ref. 50) investigated whether chromium concentrations in lymphocytes can be used as a
biomarker for long-term chromium exposure. White
blood cells accumulated significantly more hexavalent
chromium than did red blood cells. Chromium was undetectable in white blood cells, and present in only low levels in red blood cells, after administration of trivalent
chromium to rats. The investigators concluded that the
exclusive accumulation of hexavalent chromium by
white blood cells supports their use as target cells in the
development of biomarkers for assessing exposure.

Chromium: Biological Monitoring. Biological monitoring of low-level chromium exposures was studied by
two investigative groups (Refs. 36 and 41). Bonde and
Christensen (Ref. 36) examined absorption and clearance
kinetics following low-level exposure to chromium in 39
welders and 18 controls. While blood and urine chromium concentrations were significantly higher in stainless steel and mild steel welders than in controls, there
were no significant changes in chromium levels during
the work week, and urine and blood chromium concentrations did not change during a 3-week break from
welding. The investigators found these data to be consistent with a gradual buildup of chromium during longterm welding exposure.
Bukowski et al. (Ref. 41) examined factors which
could confound studies of the absorption kinetics of
chromium in workers who experienced low-level, intermittent exposures to soils heavily contaminated with
chromium. They focused on determining nonvocational
factors which could introduce errors into biological monitoring data. Males had slightly higher urinary chromium
levels than did females. Beer drinking was associated
with a significant increase in urinary chromium, while
tobacco use or exercise caused a slight decrease in urinary chromium levels. Neither welding nor engaging in
hobby activities with possible chromium exposures was
associated with increased concentrations of chromium in
the urine or blood.
Nickel. Angerer and Lehnert (Ref. 5) determined nickel
levels in red blood cells, plasma, and urine from 103
stainless steel welders who worked with chromiumnickel alloyed steel. The authors calculated that urinary
nickel levels between 30 and 50 g/L correspond to an
external exposure of 500 g/m3.
Aluminum. Elinder et al. (Ref. 59) measured aluminum
levels in bone and urine from two workers 5 years after
they had ceased routine welding of aluminum. The investigators concluded that aluminum accumulates in the
skeleton and tissues following long-term inhalation exposure and that the elimination of retained aluminum is
very slow, on the order of several years.
Zinc. Chughtai et al. (Ref. 49) showed that zinc concentrations were significantly higher in serum from gas and
arc welders than from controls.

Incidental Exposures
Degreasing Agents. Two incidents were described in
which photochemical decomposition products resulting
from reaction of degreasing agents with UV radiation
produced by GMAW or GTAW may have caused respira-

14

tory distress. In the first incident (Ref. 191), a welder developed respiratory distress and pulmonary edema after
welding stainless steel degreased with trichloroethylene.
In the second incident (Ref. 181), a welder developed severe respiratory distress and eventually died after welding metal pieces degreased with 1,1,1-trichloroethane for
3 days. Phosgene was not detected during a reconstruction of this incident at the worksite.
Coated or Contaminated Surfaces. Two cases were reported in which welders developed deficits in pulmonary
function presumably from exposure to materials generated by burning of paint or other metal surface contaminants (Refs. 119 and 185). In a third case, diffuse,
interstitial pneumonitis was shown to be related to longterm exposure to cutting oils (Ref. 162).
Allergens. Two welders were described who developed
allergic responses to chemicals incidental to the welding
process. The first welder developed bronchial asthma
while welding metal pieces contaminated with chloramine-T (Ref. 28). The second welder developed an urticarial skin reaction associated with high fever and facial
edema while welding nonstainless steel profiles filled
with polyurethane (Ref. 107).

Investigations in Animals
and Cell Cultures
Fertility
Ernst (Ref. 63) found that hexavalent chromium, administered to rats by intraperitoneal injection, causes a
reduction in testicular weight and a dose-dependent increase in the number of atrophic seminiferous tubules. At
the highest dose tested (4 mg/kg body weight), almost all
of the seminiferous tubules were completely degenerated, and there was a marked reduction in the epididymal
sperm count. Trivalent chromium did not cause these
effects.

Fibrosis
Yurui and Yu (Ref. 228) demonstrated that the proportions of collagen Types I and III change during the fibrotic process stimulated by quartz but not by welding
fumes. The authors concluded that the ratio of Type I to
Type III collagen can be used for evaluation of the fibrogenicity of respirable dusts. Hicks and Olufsen (Ref. 89)
demonstrated that myofibroblasts proliferate in fibrotic

processes in lungs exposed by intratracheal instillation to

crystalline silica or to fumes generated by GTAW of mild
steel.

In Vitro Tests
In continuing studies of the effects of hyperbaric pressure on physiological functions, Jakobsen et al.(Ref.
101) found that increasing pressure may increase the
ozone toxicity to alveolar macrophages. Using the
chemiluminescence assay, Saburova et al. (Ref. 177)
found that welding fumes did not elicit the formation of
hypochlorous acid (HOCl) by isolated PMNs. The authors concluded that HOCl does not play a role in the
pathogenic processes associated with inhaled welding
fumes.
Adamis et al. (Ref. 1) tested fumes collected from
GTAW of an Al-Mg alloy and dusts collected from other
operations in an aluminum plant in three in vitro assays
(erythrocyte hemolysis, interference with macrophage
metabolism and lysis of macrophages). While none of
the three samples was classified as hazardous, the welding fume particulates were more toxic than the other dust
samples.

Animal Studies
Naslund et al. (Ref. 143) examined pulmonary effects
of fumes generated by SMAW of black iron in sheep.
Acute exposure to welding fumes significantly increased
mean pulmonary arterial pressure and the number of leukocytes. Hematocrit values, arterial oxygen tension, and
pO2 were reduced. Chronic pulmonary exposure to the
welding fumes caused fibrosing pneumonitis and slight
emphysema.
Pokrovskaia and Cherednichenko (Ref. 163) examined the effects of fumes from five different welding
electrodes on the cardiovascular system and respiratory
tract in rats. After 1 month, there was atelectasis, swelling of the bronchial epithelium, and thickening of alveolar walls. Damaged muscle fibers were seen in the heart.
At three months, there were still substantial effects in the
lung and cardiovascular system. By 6 months, much of
the tissue damage had resolved. Fibrosis was not observed at any time.
Geleskul et al. conducted a series of studies in which
the toxicity of fumes from three different electrodes was
measured by lipid peroxidation in the liver and lung
(Refs. 71-73). Fume samples were administered to rats
by intratracheal instillation, and malondialdehyde was
determined in lung tissue or liver mitochondria. Fumes

15

from all three electrodes stimulated the formation of peroxides. Peroxidation was greatest at 1 day after treatment
in lung tissue and at 7 days after treatment in liver tissue.
In a related report by Geleskul et al. (Ref. 74), the tendency for welding fumes to cause lipid oxidation was
measured in liver mitochondria by determining the
effects on the glutathione/glutathione reductase antioxidant system.
Velichkovski et al. (Ref. 214) compared the toxicity
of fumes from four electrodes in three different assays

(chemiluminescence, hemolysis of red blood cells, peroxidation of lipids in the lung). It was concluded that
there is a strong correlation between the toxic activity of
the welding fumes in vitro and in tests performed in
vivo. Using the lipid peroxide techniques described by
Geleskul et al., (Ref. 70), Kuchuk et al. (Ref. 117) measured lipid peroxidation in liver tissue and chemiluminescence in white blood cells isolated from rats treated
by intratracheal instillation with fumes from 6 different
electrodes.

Conclusions

level exposures, and also those which are attempting to

measure small effects of exposure, are particularly sensitive to errors introduced by nonvocational exposures
and, thus, results must be viewed with caution. This is
especially true of studies such as those conducted by
Bonde on fertility in welders. Carefully planned laboratory studies in animals and epidemiologic studies in
humans are necessary to resolve questions raised by
researchers in these areas.

During this report period, important areas of research on the health effects of welding focused on the
effects of long-term exposure to low levels of chromium, and the potential effect of welding on fertility. In
addition to these studies, Sjogren et al. (Ref. 189) focused attention on potential neurologic effects associated with long-term exposure to aluminum. It is
expected that these areas will continue to receive attention in the future. Investigations of the effects of low-

17

Effects of Welding on Health IX

Section One
The Exposure

2. Fumes
Wisniewski (Ref. 223) analyzed cinematic films of
metal droplet transfer in the welding arc during GMAW
using carbon dioxide (CO 2 ) or mixtures of Ar+O 2 or
Ar+CO2 as the shielding gas. At low welding currents,
molten metal forms into droplets on the tip of the filler
wire; when the droplet diameter is larger than that of the
wire, the droplets separate and pass into the pool of molten metal. Vaporization from the droplet as it separates
from the wire can have an explosive character, originating in the neck of the drop, promoting the formation
of fumes. Processes which reduce the final growth phase
of the droplet, reduce fume formation. As the welding
current is increased to the spray mode, particles with diameters smaller than the filler wire are transported axially to the molten pool. In this phase, the droplets tend to
stay intact until they hit the workpiece and fume formation is confined to the region of the arc, which reduces
the quantity of fume. When the shielding gas is moving at
low velocities, vaporization is limited by diffusion of the
metal vapor. At higher shielding gas velocities, vaporization into the surrounding air is enhanced by convection.
Hewitt and Hirst (Ref. 87) developed a model to predict the composition of welding fumes generated by
SMAW and FCAW. Two mechanisms of transfer of
metal to the fume were considered: fractionated transfer,
in which the metals evaporate at different rates dependent on their partial vapor pressures, and direct transfer,
which results from surface depletion and spatter, is independent of melting point, and in which the metals are
present in the aerosol in the same proportions as in the
original materials. Transfer coefficients for use with the
model were developed from experiments in which different base plates were welded with flux cored mild steel
wires of varying composition. The fume composition
predicted on the basis of these transfer coefficients
closely agreed with compositions determined experimentally. In these experiments, 45% of the fume was derived
from the flux, and 39% was derived from the wire
sheath, even though the flux represented only 18% of the
total weight of the wire. Fractionated transfer accounted

1. Introduction
Welding fumes and gases originate from vaporization
of the welding rod or wire, decomposition and vaporization of the flux materials, spatter and fumes from the arc
region and weld pool, and evaporation from the molten
weld metal. All of the chemical elements found in the
welding electrode, flux, workpiece, and shielding gases
are present in the welding emissions. But, because of differences in volatility, their proportions in the fume may
not parallel their proportions in the welding consumables
and workpiece. Metals with low melting points, such as
manganese, copper, and lead, are enriched in the fume,
compared to more refractive elements such as iron, chromium, and nickel (Ref. 100).
The fume generation rate (FGR), or the total quantity
of fume emitted per unit time (e.g., gram/min), is lowest
with gas tungsten arc welding (GTAW), followed by gas
metal arc welding (GMAW), and highest with shielded
metal arc welding (SMAW) and flux cored arc welding
(FCAW). The electrode is the source of a much greater
proportion of the total emissions than is the workpiece.
Solid electrodes have the lowest FGR. Flux cored wires
have a higher FGR than do comparable covered electrodes, but flux cored wires also deposit more metal in a
given period, so that the mass of fume generated per kg
of weld (relative fume formation rate, RFFR) is comparable (Ref. 14). Helmet design, the welders posture and
position relative to the plume, and ventilation affect the
extent of a workers exposure to welding emissions (Ref.
19). The use of robotics has the advantage of removing
the operator from the immediate vicinity of the plume,
reducing exposure to the most intense fumes and radiation. However, because robotic welders are capable of
continuous operation for long periods of time, more
fumes are can be generated than with manual welding,
and requirements for exhaust ventilation may be increased (Refs. 12 and 102).
19

20

for 78% of the fume. This is consistent with the high

contribution of the flux which, being in the core of the
wire, is not accessible for direct transfer.
Hilton and Plumridge (Ref. 91) conducted an extensive series of tests of the effects of welding current and
various shielding gases on the quantity of fumes generated by GMAW of various base metals. When welding
mild or stainless steel with solid electrodes, the FGR rose
steadily as the current was increased until an unstable
transition mode was encountered at about 170 A (see
Figure 1A). With argon shielding gases, the FGR decreased or leveled out with increasing current above
170 A, as a spray mode of transfer was achieved. When
mild steel was welded with a CO 2 shielding gas, the
spray mode was never achieved, and the FGR continued
to increase with increasing current. Welding stainless
steel with a rutile flux-cored wire generated more fumes,
but less chromium, than did welding with a solid wire.
Helium-based shielding gases produced more fumes than
did argon-based shielding gases during GMAW of stainless steel, copper-base materials, or aluminum alloys
(Figure 1B). At currents above 150 A, more fumes were
generated by GMAW of an aluminum alloy containing
5% magnesium than by GMAW of an aluminum alloy
containing 5% silicon (Figure 1C). This alloy effect was
much more pronounced than was the effect of the helium
shielding gas. In general, fume levels generated by
GTAW using argon-based shielding gases were much
lower than those produced by GMAW.
Oleinichenko et al. (Ref. 153) measured the effect of
welding current and voltage on the composition of welding aerosols using flux-cored strips. At 32 V, the rate of
melting increased by 39% as the current increased from
600 to 800 A. The RFFR, measured in g/kg of electrode,
increased by 34% as the current increased from 600 to
800 A, and manganese increased by 7.5%, vanadium by
120%, and hexavalent chromium [Cr(VI)] by 10%. At a
constant current of 800 A, the rate of melting increased
only another 5% as the voltage increased from 32 to
36 V, but the RFFR increased by an additional 37%.
Olah and Pospisilova (Ref. 152) characterized fumes
generated by welding high-alloy steel with various welding techniques. SMAW with tubular wires produced the
highest FGR. Solid electrodes with a high chromium and
cobalt content had a lower FGR, but the rate of fume
generation per gram of weld (RFFR) was greater.
GMAW using wires with a high chromium and nickel
content produced less fumes, but the chromium content
of the fume was so high that chromium emissions were
almost equal to those from SMAW. GTAW using wire
high in chromium and nickel generated about one-tenth
the quantity of fumes produced by SMAW. Although the
fumes contained up to 23% chromium, GTAW still pro-

duced the lowest quantity of total chromium of all the

processes studied.
Medack and Heinze (Ref. 135) measured and characterized the emissions from GMAW of steel coated with
zinc, tin, and aluminum. Four to six times more fumes
were generated by welding metal sheets coated with zinc
and tin than by welding aluminum-coated and uncoated
steel. When the emissions associated with welding these
coated steels were compared with allowable workplace
exposure limits, zinc was the limiting factor for zinccoated steel, for tin-coated steel it was lead, and for aluminum-coated or uncoated steel it was manganese which
originated from the electrode.
Mori et al. (Ref. 140) measured the chromium and
nickel generated by SMAW with three Japanese fluxcoated electrodes containing 9 to 13% chromium and 39
to 68% nickel and by submerged metal arc welding with
a nickel alloy wire. The base metal was mild steel containing no chromium or nickel. Fumes were collected at
12 sampling points arranged in a three-dimensional matrix at three heights directly above the arc and, at the
same three heights, displaced 50 cm from the front, to the
rear, or to one side of the arc. SMAW produced more
fume than did submerged arc welding. Using an electrode with 12.2% chromium and 68.1% nickel, the fume
concentration was 254 mg/m3 and the chromium content
was 5.3% at a height of 50 cm directly above the weld.
At 150 cm and 250 cm directly above the weld, the fume
concentrations were 68 and 65 mg/m3, respectively, and
the chromium concentrations decreased to 4.3%. Fume
concentrations were dramatically lower in samples collected at the side and in front of the arc, and chromium
was only detected in one of these samples. Directly
above the arc, nickel concentrations decreased from
0.54% at 50 cm to 0.25% at 150 and 250 cm. Submerged
arc welding of the nickel wire produced fume concentrations of 118, 13, and 6.6 mg/m3 at 50, 150, and 250 cm
directly above the weld, respectively. Notably, the percentage of nickel in the fume from the submerged arc
welding process increased with sampling height above
the weld, which was attributed to the low density of
nickel oxide compared with iron oxide.
Eichhorn and Nies (Ref. 58) evaluated a variant of the
narrow-gap submerged arc welding process with two
wire electrodes. The system used smaller and more easily manipulated welding heads and was purported to be
more efficient and flexible than conventional single wire
electrode techniques. Large quantities of fumes were
generated in tests with a barium-containing welding
powder, used to increase the shrinkage of the slag during
cooling. While this is unlike conventional submerged-arc
welding which is characterized by extremely low fume
emissions, the fumes were still considerably less voluminous than those formed during FCAW with barium-

21

Figure 1AFume Generation Rate During GMAW of Mild Steel Using Solid Mild Steel
Wire Showing the Effects of Increasing Concentration of CO2 in the Shield Gas
with Increasing Current. Hilton and Plumridge, Ref. 91

Figure 1BFume Generation Rate During GMAW of Stainless Steel Using Solid Stainless
Steel Wire Showing the Effects of Helium in the Shield Gas with Increasing Current.
Hilton and Plumridge, Ref. 91

Figure 1CFGR for GMAW of Aluminum Alloys Containing Either 5% Magnesium

(Alloy No. 5636) or 5% Silicon (Alloy No. 4043) Showing Effect of Magnesium in the Metal
and Helium in the Shield Gas with Increasing Current. Hilton and Plumridge, Ref. 91

22

containing flux cored electrodes. In addition, barium accounted for 12% of the fume with the narrow-gap submerged arc process, compared with up to 32% of the
fume from flux cored welding. The investigators concluded that local ventilation would be capable of preventing barium concentrations from exceeding the
workplace limit (0.5 mg/m3).
2.1 Analytical Techniques. Ukkonen et al. (Ref. 206)
described a method for determining the mass median diameter of an aerosol by combining data from two realtime aerosol monitors (a photometer and an electrical
aerosol detector). The method worked well for particles
with densities close to unity (tobacco smoke and dioctyl
phthalate). However, because of their high density, estimated by Ukkonen et al. to be 3 g/cm3, the method did
not accurately determine the diameters of welding fume
particulates.
Battistoni et al. (Ref. 17) used an aerosol photometer
to measure particle concentrations at points 0.3 and
1.8 m above a Migtrain (TM, Kobe Steel Ltd.) automatic
welding system (GMAW) using a mixture of 80% Ar and
20% CO2. From the two measurements, an overall emission rate of 9.6 mg/min was calculated using a two-point
eddy diffusion model.
Particles in welding fumes are formed by condensation of vapors of metal oxides to form condensation
nuclei which further interact with each other and with
uncondensed vapors to form larger particles. During this
process of particle-building, the concentrations of the
various components of the particle may not be homogeneously deposited throughout the depth of the particle,
and certain elements may be enriched on the surface
while others may be present in higher concentrations inside the particle. Goschnick et al. (Ref. 80) used secondary neutral mass spectrometry (SNMS) to analyze the
composition of particles generated by SMAW with a
stainless steel electrode. Chromium in the fume particles
was enriched by a factor of three relative to iron, based
on the original composition of the electrode. Removal of
successive layers by argon sputtering revealed no variations in the chemical composition between external and
internal layers of the particles. This contradicts earlier
work by Grekula et al. (Ref. 81) and Malmqvist et al.
(Ref. 129) who, using scanning electron microscopy and
energy dispersive X-ray analysis, showed differences in
the chemical content of the surface and core of particles
generated by SMAW of stainless steel.
Bauer et al. (Ref. 18) compared two methods for determining Cr(VI) in fumes from GMAW and SMAW of
stainless steel. With the first method, area and breathing
zone samples were collected on polytetrafluoroethylene
(PTFE) filters. The PTFE filters were eluted with alkaline buffer solution and Cr(III) and Cr(VI) were captured
by anion exchange resins. In the second method, fume

samples were collected on glass fiber filters. The filters

were placed in an absorption medium in a PTFE flask,
rinsed with distilled water, filtered through a 0.2 m cellulose acetate filter, and acidified with nitric acid and
ammonium nitrate. For both methods, Cr(VI) was determined by AAS. The results of the two methods were
comparable for both personal and area samples. The
samples could be preserved in alkaline carbonate for 3
days without affecting the Cr(VI) measurement.
2.2 Analysis of Metals in Biological Tissues. Identification of metals in biopsied or autopsied tissues is sometimes useful in discerning the cause of disease. Seemann
et al. (Ref. 180) investigated whether trace elements can
be reliably determined in lung tissue specimens that have
been fixed and stored in formalin, because occasionally
only fixed tissues are available for examination. In this
work, the chromium and nickel contents of eight postmortem lung specimens were determined in the lungs
from four men who had died of lung cancer between
1988 and 1989. One of these men had been employed as
a plumber and did soldering and welding for 40 years.
Two of the other men had been coal miners, and the occupation of the third man was not specified. Chromium
and nickel concentrations were determined by flameless
atomic absorption spectrometry (AAS) after lyophilization, wet ashing under pressure, chelating, and extraction. The concentration of chromium and nickel was
markedly higher in the lung tissue from the welder than
from the other three men. The investigators concluded
that their method provides a reliable means of determining trace metals in fixed lung tissue.

3. Workplace Exposure Limits

Balchin (Ref. 12) reviewed the British health and
safety regulations as they apply to welders, noting that
standardization of these regulations within the European
Community (EC) was being anticipated by adopting, in
part or in whole, the ECs European Norms as British
Standards. Sutton (Ref. 194) evaluated the intrinsic hazards of substances used, or arising from the welding of
stainless steel with respect to the British Control of Substances Hazardous to Health Regulations of 1989. He observed that in fumes from SMAW of stainless steel,
hexavalent chromium [Cr(VI)] is usually the hazardous
component that determines the requirements for air
cleaning or personal protection. The Cr(VI) content of
the fumes produced by GMAW and GTAW is so much
lower than the Cr(VI) content of fumes from SMAW that
the standard for total chromium of 0.5 mg/m3 can usually
be applied.
Dryson and Rogers (Ref. 55) conducted a survey of
airborne contaminants in the breathing zone of 16 weld-

23

ers performing SMAW, GTAW, GMAW, and plasma cutting in seven New Zealand industrial plants. Samples
were collected inside the welding helmets for 2 to 4
hours, during that part of the workshift devoted primarily
to welding. Urine samples were taken at the end of the
shift on the same day as the breathing zone samples.
Total dust levels were higher for SMAW, plasma cutting,
and GMAW than for GTAW. Levels of ozone, fluoride,
and zinc were not greater than 20% of the New Zealand
workplace exposure standards in any of the samples collected. Of the 16 welders in the study, six were exposed
to excessive levels of at least one pollutant. Allowable
limits for total fume, nitrogen dioxide, and chromium
were occasionally exceeded. Analyses of urine did not
show excessive absorption of nickel or chromium.
Van der Wal (Ref. 213) analyzed fumes and gases collected in area samples and in breathing zone samples
from welders working at 18 different plants and locations
in Holland. All of the work was performed in large
rooms without local exhaust ventilation. Data from previous studies (Ref. 211 and 212) were included in the
analyses. Figure 2 shows the distribution of fume concentrations in the breathing zone generated by different
welding methods. GTAW was the only welding process
for which there were no measurements of breathing zone
fume concentrations exceeding the Dutch occupational
health standard (TLV) of 5 mg/m3. SMAW and GMAW
of stainless steel produced Cr(VI) levels above the Dutch
TLV (0.05 mg/m3) as did polishing of stainless steel, but
plasma welding or cutting of stainless steel did not produce excessive levels of total chromium or Cr(VI). Carbon monoxide (CO) levels were below the Dutch TLV of
50 ppm, except for metal spraying operations, where
they ranged from 40-160 ppm. No measurements of NO2
exceeded the Dutch TLV of 5 ppm. Ozone concentrations did exceed the Dutch TLV of 0.1 ppm for GMAW
of stainless steel and aluminum, FCAW of mild steel,
and plasma cutting of Inconel alloy.
Withers et al. (Ref. 224) used Particle-Induced X-ray
Emission (PIXE) to analyze elements in fumes collected
inside the helmets of welders during spot welding of
zinc-plated steel sheets. Relative concentrations of chromium, manganese, iron, copper, zinc, and nickel were
determined. Iron was the predominant metal found in all
but one of the samples. The proportion of zinc, the second most common metal, varied from 7% to 48.5%.
While repeated analyses of a single sample showed good
reproducibility, there was a large variability in the relative concentrations of elements among different samples.
Because of this, Withers et al. (Ref. 224) concluded that
no single element can be used as an indicator for the concentrations of other elements in a sample and a monitoring program based on the analysis of one reference
element may be ineffective.

Paul (Ref. 161) measured lead exposures during welding and cutting operations in 130 automobile body repair
shops in Dresden, Germany. The welding fume concentrations in the workplace air exceeded the German Democratic Republic (GDR) standard of 5 mg/m3 in more
than 50% of the measurements taken during welding operations and 90% of those taken during cutting operations. During welding of car bodies, lead levels were
rarely in excess of the GDR standard of 50 g/m3, but
fume concentrations from cutting operations exceeded
the standard more than 50% of the time. The high environmental lead levels in these body repair shops were attributed to lead used in car finishes and lead in street dust
deposits on the car bodies.
Gorban et al. (Ref. 78) investigated emissions from a
series of Russian electrodes used in welding of iron and
stainless steel without preheating. The emissions of NO2,
CO, and hydrogen fluoride (HF) from these electrodes
were insubstantial. They measured the FGR and the
RFFR for iron, manganese, nickel, copper, silicon, and
soluble and insoluble fluoride, but not for chromium, and
calculated the volume of air necessary to dilute the emissions of the most critical component from each electrode
to a concentration below the workplace standards in the
Soviet Union. They concluded that the workplace standards could be achieved using these electrodes in a wellventilated workplace.
Zaks (Ref. 229) evaluated published data on the emissions from several different covered electrodes in terms
of allowable workplace limits. Among the coatings evaluated, total fume emissions were generally greatest with
acid coatings, least with rutile coatings, and intermediate
with ilmenite coatings. Manganese, HF, CO, and nitrogen pentoxide were measured, and the volume of air that
would be necessary to dilute all of the measured emissions to below the Soviet Union workplace limits was
determined. Manganese was the governing emission in
the case of acid ilmenite, rutile, and cellulose coatings,
but fluoride was the critical emission for some of the
electrodes with basic coatings. For stainless steel electrodes, the Cr(VI) emission rate varied from about 0.15
to 1.0 grams chromate per kg electrode, with the median
content of 0.4 g Cr(VI)/kg electrode (Ref. 230). Zaks
concluded that there is a large variation in emission characteristics reported by different investigators, even for
the same types of electrodes.

4. Ozone
Ozone is generated from atmospheric oxygen by the
action of ultraviolet (UV) radiation. GMAW and GTAW
of aluminum generate the highest ozone levels, but
GTAW and GMAW of stainless steel can also generate

24

Figure 2Cumulative Frequency Distribution of Welding Fume Concentrations in the

Breathing Zone of Welders During Welding without the Use of Local Exhaust Ventilation.
(MMA-SS = SMAW of Stainless Steel; MIG-SS = GMAW of Stainless Steel; MIG-Al = GMAW
of Aluminum; MMA-unall = SMAW of Unalloyed Steel; MIG/MAG unall = GMAW of
Unalloyed Steel; SAW-unall = Submerged Arc Welding of Unalloyed Steel;
TIG-SS = GTAW of Stainless Steel). Van der Wal, Ref. 213

substantial amounts of ozone, particularly with argon

shielding. Ozone is highly reactive and is rapidly consumed by reaction with other components of the welding
fume. Thus, welding processes that generate large quantities of fumes (SMAW and FCAW) are not generally associated with high exposures to ozone. Conversely, a
reduction in the FGR may result in increased UV radiation and, consequently, increased ozone formation (Ref.
14). The highest concentrations of ozone are normally
found within 50 cm of the arc. However, because the
path of UV radiation can sometimes allow it to escape
absorption by the fumes, ozone may be generated at a
short distance from the arc, but outside the plume of
welding fumes, thereby, escaping reaction with fume
components (Ref. 194).

Brehme et al. (Ref. 38) used a chemiluminescent

ozone monitor to simultaneously measure up to 16 ozone
concentrations in the welding plume and in the breathing
zone of welders. The tests were conducted during various gas shielded welding operations in three German
manufacturing plants. In all of the plants, the average
concentrations of ozone in the breathing zone were well
below the workplace limit of 0.1 ppm. Ozone concentrations were lowest in an automobile plant using argonshielded GMAW, and highest in a chemical apparatus
plant where GTAW was performed with a 100% argon
shield. In a machinery manufacturing plant, breathing
zone ozone concentrations were slightly higher when
ventilation was used than when it was not, indicating that
the ventilation system was improperly designed. In ap-

25

plications involving aluminum or aluminum/magnesium

electrodes, the ozone concentrations often exceeded the
0.1 ppm workplace limit without ventilation and sometimes reached the allowable workplace limit even with
the use of local ventilation.
Olsson et al. (Ref. 156) found that a shielding gas
containing Ar plus 8% CO2 is optimum for GMAW at
welding currents greater than 150 A. They recommended
inclusion of 0.08% nitric oxide (NO) to convert the
ozone to O2. This reaction also converts the NO to NO2,
which they state is less harmful than ozone to the welder.

5. Electromagnetic Radiation
Welders may be exposed to radiation from their own
arc or from the arc of nearby welders. A common occasion of inadvertent exposure occurs when striking the arc.
Since welders are often unable to see through the lens in
the welding helmet, to position the electrode they must
raise the face piece and then flip it down with a head
motion as the arc is struck. Mistiming of this maneuver
can result in exposure of the eyes to intense radiation
from the arc (Ref. 198). According to Bauer (Ref. 20),
protection from this type of injury is now available in
welding helmets with lenses that darken on arc ignition.
Barth et al. (Ref. 15) monitored the exposure of nonwelders (bystanders) to UV generated by GMAW with a
CO2 shield. The UV exposures of nonwelders, recorded
with polysulphon films, were 8 times higher than the
limits recommended by the National Institute for Occupational Safety and Health (NIOSH). Clinical examinations showed that chronic damage of the external parts of
the eyes occurred more than twice as frequently in bystanders than in controls. Chronic conjunctivitis and
pinguecula (yellow spots on the exposed conjunctival
bulbi) were the most common outer eye symptoms observed among the bystanders. These results indicate the
need for the use of shields to protect persons working in
the vicinity of welding job sites from UV radiation.
The use of lasers in welding applications creates another occasion for exposure to UV radiation. Injury to
the eye can occur not only from directly viewing the
beam, but also from exposure to reflected laser radiation.
UV lasers are particularly hazardous because their beam
is invisible (Ref. 95).
Lasers are classified into four hazard categories, with
class IV being the most dangerous. The majority of industrial lasers used in manufacturing are Class IV and,
thus, special safety features are required to prevent exposure. Class IV lasers present a radiation hazard to the
skin and eyes of unprotected operators or bystanders.
With infrared (IR) laser beams, visible light and UV radiation are released during laser/target interactions. In se-

lecting eye protection, it is important to consider not only

scattered laser radiation, but also noncoherent UV radiation and blue light. Heat-resistant clothes and gloves
must be chosen to resist the scattered IR laser beams
(Ref. 90).
Industrial laser systems package the laser inside an
enclosure designed to attenuate the accessible radiation
to a Class I level. Safety features include a system of interlocks which shut down the entire system should hazardous conditions develop. Other safety features of the
work station include magnified viewing of the work area
provided through a closed circuit television, or microscope viewing of the workpiece. Also, the use of optical
fibers to deliver the laser radiation to the workpiece affords more control of the laser beam (Ref. 109).
The standard welding curtains used in conventional
welding are not useful for laser welding because they
cannot withstand the high energy of the laser beam. An
effective laser barrier should both identify locations of
laser work areas and impede transmission of laser beams
outside the welding area, preventing exposure of unprotected workers. Another useful function of the barrier is
to allow the detection of stray laser beams. The laser barrier curtain design described by Wilson (Ref. 220) includes a simple barrier coating that is visually marred
upon exposure to stray laser beams. The location of
marks on the barrier can indicate improper beam position, variation in beam intensity, or inappropriate weld
technique. Any visual changes in the curtain would then
indicate that unsafe welding conditions are, or were,
present. Marks on the barrier could also be used to train
workers in proper laser application techniques.
Hazards associated with the use of laser technology
include not only the potential for exposure to radiation,
but also electrical hazards, hazardous chemicals required
for laser operation, and toxic gases and particles generated during the use of the laser. As with conventional
welding, emissions must be properly removed to eliminate exposure to toxic chemicals (Ref. 109).
Engel et al. (Ref. 60) described research conducted in
the Federal Republic of Germany using a pilot installation developed to study the emissions from cutting and
welding with lasers. Studies of emissions from cutting
steel with CO 2 lasers showed that more fumes are released when cutting stainless steel than nonalloyed steel.
The particle size range was very narrow in emissions
from all steels; 87% of the particles from galvanized
steel, 85% of the particles from stainless steel, and
76% of the particles emitted from nonalloyed steel were
between 0.021 m and 0.18 m in diameter. Elemental
analyses indicated that the composition of the particles
is similar to that of the parent metal. With galvanized
steel, the proportion of iron increased, and that of zinc
decreased with the thickness of the metal being cut.

26

Electron microscopic analyses showed that the particles

exist as individual spheres and as agglomerates of spherical particles.
Hietanen and von Nandelstadh (Ref. 90) evaluated
hazards associated with optical radiation generated during the operation of lasers in metal industry applications.
The intensities of both the laser beam and scattered laser
radiation were determined. During normal work shifts,
the diffuse reflection of the laser beam was small. The
maximum irradiance measured was about 100 W/m 2 ,
compared with the maximum permissible exposure of
1000 W/m2. Maximum permissible exposure times for
unprotected eyes during normal operations of the tested
laser were calculated to be 95 seconds for CO2 welding
of stainless steel and 124 seconds for carbon steel.

6. Electromagnetic Fields
Electromagnetic fields (EMFs) are produced by electric arc welding power sources. Zyubanova et al. (Ref.
232) calculated the intensity of the magnetic field created
by current flowing through electric arc welding equipment. They showed that exposure to magnetic fields
could be minimized by increasing the distance between
the welder and the electric wires and by keeping the distance between the power cables to a minimum (e.g., by
using coaxial cable).

7. Incidental Exposures
7.1 Production Coatings. A laboratory study of thermal
degradation products of paints used on steel in the
Finnish ship building industry was conducted by
Henriks-Eckerman et al. (Ref. 84) to develop a simple
and representative sampling procedure for the workplace. Pollutants generated by heating painted steel
plates to about 350o C were collected with an array of
sampling devices (adsorption tubes, bubbler absorbers,
and filters) and analyzed by high-pressure liquid chromatography and gas chromatography/mass spectroscopy
(GC/MS).
Low molecular weight aldehydes and acids were emitted from all paints. The most abundant degradation products usually originated from the paint binders and
included phenol and bisphenol-A from epoxy paints, aliphatic organic acids and phthalic anhydride from alkyd
paints, and butyraldehyde and butyric acid from polyvinylbutyral paint. Toluene, xylene, ethylbenzene, propanol, butanol, methyl isobutylketone and cyclohexanone
were detected and were thought to arise from the paint
solvents. Hydrogen chloride (HCl) was emitted from
chlorinated rubber. Low molecular weight aldehydes, acids, and solvents were also detected. Ethanol was a typi-

cal emission from ethyl silicate shop primers which

emitted the least irritating degradation products. On the
basis of this work, the investigators recommended a sampling procedure for on-site studies which included a
chemosorbent tube for aldehydes and a tube of adsorbent
resin preceded by a glass fiber filter for particulates and
condensable organic chemicals.
Using this sampling procedure in a Finnish shipyard,
Engstrom et al. (Ref. 61) studied the exposure of metal
workers to paint degradation products generated by
welding, flame cutting, and straightening painted steels.
With the exception of the higher molecular weight aliphatic acids, which may have been more completely degraded in the workplace, the same degradation products
were identified in the field as in the laboratory. Straightening processes produced the lowest quantity of total
fumes but generated organic pollutants in amounts equal
to those produced by the other processes. This indicated
a special need to monitor organic compounds and to not
rely on measurements of total fume when evaluating
risks associated with the straightening process.
Concentrations of individual compounds were usually
below allowable limits. Even so, there were some complaints of irritation, suggesting that effects were additive.
The investigators suggested changes in the design process to reduce the risk of exposure to irritants and organic compounds generated from paints on surfaces to be
welded. The use of ethyl silicate shop primers would
substantially reduce exposures to irritating degradation
products. Areas to be welded should be left bare whenever possible by procedures such as masking with tape
before painting or by removal of paint before welding. In
some special types of work, such as flame cutting, which
generate large quantities of organic pollutants, special
respiratory protection (e.g., air supplied helmets or respirators) may be necessary.
A health hazard evaluation was undertaken by
NIOSH in response to a request by an employee who became light-headed and nauseated on three occasions during the computerized, semi-automated plasma arc cutting
of polyvinyl chloride (PVC)-coated galvanized steel
sheets (Ref. 6). The investigators demonstrated that,
while little or no fume was visible during the cutting of
the uncoated sheets, a large, spreading plume of smoke
was generated by cutting the PVC-coated steel sheets.
7.2 Radiation Exposure. Onodera (Ref. 157) characterized the aerosols generated during the decommissioning
of a nuclear power plant in Japan. Cutting 60Co-contaminated stainless steel pipe with a plasma torch generated
fumes at a much greater rate than cutting with a band saw
or reciprocating saw. However, 30 times more radioactivity was released per cut with the reciprocating saw
than with the plasma torch. Cutting the reactors internal
stainless steel components under water markedly re-

27

duced emissions of fumes and airborne radioactivity. Radioactive emissions were 100 to 1000-fold higher when
cutting the stainless steel clad reactor shell with an oxyacetylene torch in air than when cutting with a plasma
arc underwater.
Carmichael and Haynes (Ref. 46) described the emissions of tritium (a radioisotope of hydrogen) encountered
during the repair of distillation columns used for extracting tritium from tritiated water at a Canadian nuclear
power plant. Tritium and tritium oxide were emitted
while cutting the columns with a tube cutter, welding the
tube ends, or heating the cut area with a propane torch.
Grinding after welding did not introduce further radioactive contamination but may have enhanced its spread.
The use of rubber gloves and plastic suits, limiting work
periods to one hour, and isolation of the tritium-emitting
work in plastic tents with local area ventilation prevented
measurable tritium uptake by the workers, whose exposures were monitored by measuring levels of radioactivity in urine.

8. Hygiene and Work Practices

8.1 Ventilation and Air Cleaning Equipment. Workplace exposure to welding fumes may be reduced by natural or forced air dilution, by exhaust hoods at fixed
locations, by fixed or portable local exhaust systems, or
by exhaust systems mounted on the welding gun (Ref.
21). Capture of welding fumes as close to the source as
possible is most desirable. Vidmar (Ref. 216) surveyed
available equipment and provided guidelines for its design and selection. Fixed hoods may be used for repetitive welding in the same location, but movable hoods or
portable fume/dust collectors, either connected to a central air cleaning system or equipped with intrinsic air
filters, are necessary when welding is performed in
multiple locations or with varying welding gun positions.
Centralized collection systems, used when source capture is not possible, are less efficient and require cleaning
of much larger volumes of air.
Three types of particle removal devices are used with
welding fume extraction systems to prevent exhaust air
laden with welding fumes from being released into the
atmosphere. Electrostatic precipitators can capture very
fine particulates. Bag filters and cartridge filters can handle a wide variety of pollutants. While bag filters have
been the predominant fume collection method, they are
being replaced by cartridge filter systems, which can
capture large amounts of particulates in a small area. Removal of odors or organic vapors is usually accomplished with these systems by passing the filtered
exhaust air through a bed of granular activated carbon
(Ref. 216).

Tum Suden et al. (Ref. 205) used computer simulations to estimate the capture efficiencies of eight different
welding hood configurations for SMAW. Their results
correlated well with breathing zone particle concentrations determined experimentally using a welding fume
simulator developed by the American Welding Society
(Ref. 4). The flow rate through the hood, the hood aspect
ratio, and the welders position relative to the hood significantly affected the breathing zone fume concentration.
Jakubcik (Ref. 102) described ventilation and air filtration schemes in a large welding shop in Czechoslovakia. Improved ventilation when tack-welding long
pressed pieces was achieved by enclosing the piece in a
ventilated manifold made of removable sections which
can be opened at the section where welding takes place.
Jakubciks study also demonstrated that a large mobile,
multi-stage filter could maintain welding-generated pollutants in workplace air below regulatory limits even
when conducting GMAW with a robotic welder using a
2.5-mm wire with an FGR of 19 mg/sec.
Van der Veen and Regensburg (Ref. 210) designed an
ergonomic welding table designed to reduce the static
loads on the arm muscles and alleviate uncomfortable
working positions while at the same time reducing some
of the hazards associated with welding. Three sides of
the table slope backwards to guide thermal flow away
from the welder. Built-in ventilation removes 90 to 95%
of the welding fumes and gases. The interior surfaces are
coated with UV-absorbing materials to block exposure to
indirect radiation, and the height of the table was made
adjustable to allow the welder to assume the most comfortable position relative to the workpiece.
Cornu and Muller (Ref. 51) used helium as a tracer
gas to measure effectiveness of gun-mounted fume exhaust devices. They concluded that the position of the
exhaust vents in relation to the fume being generated is
very important in determining collection efficiency. Several Russian studies described fume collection and air
cleaning devices. Romanenko (Refs. 172 and 173) and
Gorban (Ref. 79) evaluated small-scale fume collectors
equipped with filtration devices. Butenko (Ref. 45) evaluated a small-scale filtration device, and Efimov (Ref.
57) evaluated a large hood to be used with automatic
welders.
8.2 Protective Gear. Steinegger and Walti (Ref. 193)
evaluated two systems which used activated carbon filters to remove ozone from the breathing zone during
GMAW. The first was a half-face mask, placed over the
nose and mouth; the other used a pump to provide filtered air inside the welding helmet. The efficacy of these
devices was assessed during GTAW of AlMgSi1. Both
systems reduced the ozone concentration in the breathing
zone at least tenfold, and it was concluded that either

28

mask could be used to conform with allowable workplace limits for ozone.
Alekseeva et al. (Ref. 3) tested seven sets of masks
designed for protection from welding aerosols and intense visible light. Particles smaller than 0.4 m penetrated the filters more readily than particles between 0.5
and 0.9 m in diameter. The resistance to breathing was
judged to be negligible as there was only a slight difference in pressure inside and outside the mask. The optical
filters were judged to be inadequate by welders who
wore them while conducting SMAW, GTAW, and
GMAW. Manz (Ref. 131) noted that welding helmets
with filter plates provide protection against radiation,
sparks, and spatter, but are not intended to protect against
larger flying fragments generated by chipping, grinding,
and polishing. Goggles or glasses with side shields are
generally necessary to provide eye protection during
these operations.
Jung (Ref. 105) examined the protection and durability offered by heat- and flame-proof clothing. They
found that some materials lost their resistance to ignition
after 10 to 20 laundering cycles. This was true even for
clothing treated with a commercial product guaranteed to
provide permanent protection against flames. The heat
protection provided by thirty materials was tested and ten
were found to be superior to asbestos.
8.3 Accidents. Reports on selected fatal accidents that
occurred at construction sites between 1979 and 1982
were analyzed by Trent and Wyant (Ref. 203). Of the 20
reports of fatalities that involved welders, ten were due
to falls or to being struck by falling objects. Seven were
due to environmental hazards such as electric lines, combustibles, explosive substances and atmospheres, and
pressurized lines and vessels; two resulted from contact
with the power or fuel supply for the welding apparatus;
and the remaining one resulted from contact with the
welding apparatus itself. The authors concluded that
most of these accidents were preventable.
Three fatal accidents in which welders died as a result
of falls at construction sites were described by NIOSH in
1990 and 1991 (Refs. 146, 147, and 148). In two of the
incidents, the welders were wearing safety belts and lanyards, but neither worker was tied off. In the third incident, the welder either tripped or stepped backwards into
an 18 by 24 inch skylight opening and fell approximately
20 feet.
Confined spaces are particularly dangerous because
they can harbor invisible hazards, such as explosive
gases or toxic chemicals, and may be deficient in oxygen. Oxygen depletion in a sealed chamber can result
from rust formation or bacterial growth. While infrequent, accidents in confined spaces are often fatal. Unsuccessful rescue attempts, made by untrained persons
unaware of hazards associated with confined spaces,

sometimes result in multiple fatalities. To illustrate the

dangers of such conditions and to stress the need for
properly training workers who may encounter confined
spaces, Rekus (Ref. 169) reviewed three accidents involving welders. The first occurred on a barge. A worker
entered a cargo hold which had been sealed for 10 years
and was overcome by insufficient oxygen. He was observed floating face down in the water in the bottom of
the hold by two other workers who successively entered
and were also overcome by lack of oxygen. Finally, a supervisor trained in rescue techniques arrived at the scene,
donned a self-contained breathing apparatus, and removed the three men from the hold. The two earlier rescuers died and the original entrant eventually recovered
consciousness, but he suffered brain damage.
In a similar accident, which took place on an oil drilling rig, a welder cut open and entered a tank which had
been sealed for several years. The welder overcome, presumably, by lack of oxygen, fell face down in the muck
at the bottom of the tank and drowned before he could be
rescued. Two men, who entered the tank in an attempt at
rescue, were also overcome but were later revived by coworkers. In the third case reviewed by Rekus (Ref. 169),
a mechanic and a welder working inside a stainless steel
chemical tank used plant air for ventilation. Unfortunately, this air contained only 10% oxygen. Normal air
contains 21% oxygen, and the minimum safe oxygen
concentration is 19.5%. The two workers collapsed from
insufficient oxygen and were not rescued in time to prevent their deaths. The standby man who entered in an
attempt at rescue was also overcome, but he regained
consciousness and escaped unharmed.
Rekus (Ref. 169) concluded that the principal safety
and health problems from welding in confined spaces are
physical hazards, such as those discussed by Trent and
Wyant (Ref. 203), as well as those peculiar to tanks and
other enclosed spaces. Paramount among his recommendations was evaluation of the atmosphere inside a confined space from the outside, using instruments to detect
oxygen levels and toxic or combustible gases, before attempting to enter.
Shaikh and Bhojani (Ref. 182) surveyed occupational
injuries among 36 oxyacetylene torch welders engaged
in roadside repair of automobile mufflers in Pakistan.
The welders had suffered a total of 340 injuries during
the three months prior to the survey. Of these, 124 were
caused by foreign bodies in the eye, 55 were facial injuries, and 161 were burns. Only 16 of the welders wore
goggles, and no other protective devices were used.
Accidents in which the worker is struck or crushed by
the robot arm are the major hazards associated with robotic welding. While performing welding operations, the
robotic welding arm does not typically move at rapid
speeds. However, the arm may move at a rapid and

29

dangerous speed in nonwelding maneuvers, such as positioning and cleaning the gun. To prevent such accidents,
it is essential that personnel be kept out of the robots
working envelope. On occasions which necessitate entry
into the robot enclosure (i.e., for cleaning, maintenance,
reprogramming and adjustment), a combination of failsafe design and administrative controls must be put in
practice to assure that the robot is immobilized. Mangold
(Ref. 130) described the various types of robot systems
which are in use for welding and categorized them according to level of risk. The majority of welding robots
in current use are articulated electric drive arc welders.
Gantry robots, which have very large work envelopes,
bear the greatest potential for accidental injury.
Bussenius (Ref. 44) reviewed the dangers to welders
and other workers repairing and maintaining tanks containing combustible liquids and gases. He developed formulas predicting rates of loss of fluids from pressurized
and unpressurized tanks. Safe working distances were
then calculated based upon the rates of loss of the fluids
and their flash points.

manual workers which included a large number of welders. Problems of work organization (monotony, accelerated pace, restlessness, and fear of errors) were
statistically (p = 0.001) related to stress among these
workers, as were ergonomic factors. A significantly
smaller fraction of welders reported experiencing stress
than did clerical employees in the same factory and
shipyard.
In a survey of occupational stresses conducted in
Sweden in 1990, the mental stresses associated with
welding were rated very high compared with over 100
other occupations, while physical stresses were rated
only slightly greater than average. A finding of high
work load with low self-determination (little opportunity
to influence ones own work situation) was also greater
than average among welders (Ref. 75). Gerhardsson
(Ref. 75) noted that this combination of factors has been
linked to increased incidence of cardiovascular disease,
nervous symptoms, and fatigue.

8.4 Stress. Valente and Chiapperini (Ref. 209) measured

the pulse rate, blood pressure, oral temperature, and perspiration rate of four welders involved in fabrication of a
locomotive. In addition to these physiological parameters, they also evaluated the microclimate in which the
welders worked, measuring wet bulb temperature and the
surface temperatures of the welders hands and tools.
They found that the welders were not subjected to excessive physical stress, but concluded that welding risks are
increased by hot working conditions, leading to increased metabolic rates among welders.
Richter (Ref. 170) evaluated the physical and psychological stress experienced by 12 electric arc welders in a
large chemical plant in the GDR. Pulse and respiratory
rate were measured throughout the shift and recorded by
telemetry. Each welder also wore a microclimate monitor
under his clothing which recorded skin temperature and
relative humidity. These measurements were correlated
with observations of the welders work activities. The
pulse rates of the two welders with the highest pulse rates
were greater than 120 during the work shift, which corresponded to a GDR work hygiene rating at which adverse
health effects are almost certain.
Psychological stresses were attributed to the necessity
to exercise extreme caution during welding, the meticulousness of the work, lack of communication with coworkers, lack of visual distractions, and the repetitive nature of the work. Physical stresses included extremes of
air temperature and humidity, awkward work postures,
lengthy contractions of muscle groups, and exposure to
fumes, gases and UV radiation.
Hyytiainen and Uutela (Ref. 97) studied work factors
related to psychological stress among a group of Finnish

Section Two
Effects of Welding
on Human Health
9. Respiratory Tract
9.1 Pulmonary Function and Bronchitis. Tests of lung
function are frequently used as measures of the health
status of the lungs. Many of these tests can be performed
with a simple instrument, the spirometer, which measures volume and flow rates of exhaled air. Forced vital
capacity (FVC) is the maximum volume that can be exhaled after a maximal inhalation. The volume exhaled
with maximum effort in one second is called the forced
expiratory volume (FEV1). The FEV1 is normally about
80% of the FVC. In restrictive lung diseases, such as pulmonary fibrosis, both the FVC and the FEV 1 are reduced, but the ratio between the two (FEV 1 /FVC) is
normal or slightly increased. In obstructive lung diseases, such as bronchial asthma, the FEV1 is reduced to a
greater extent than the FVC, giving a lower than normal
ratio of FEV 1 /FVC. Other frequently measured lung
function parameters are forced expiratory flow rates
(FEF), FEF from 25 to 75% of FVC (FEF25-75), and terminal flow (FEF75-85) which are indicative of impairment
in the small airways. Subnormal or declining values of
Peak expiratory flow rates (PEFR) are regarded as symptomatic of asthma. These tests are rarely used for definitive diagnosis of disease in patients, but they are useful
for following the progress of disease or for epidemiologic assessments of groups of workers exposed to

30

known or suspect pulmonary irritants. The contribution

of welding fumes and gases to the development of pulmonary function deficits and respiratory disorders such
as bronchitis remains uncertain, and conflicting results
have been reported by different investigators.
In 1979, Cotes et al. (Ref. 52) published the results of
a cross-sectional survey of respiratory symptoms and
lung function in 609 shipyard workers from northeast
England. The incidence of chronic cough and breathlessness on exertion was related to both current smoking and
the proportion of time spent working in confined spaces
as a welder or caulker/burner. Lung function impairment
among nonsmokers was related to employment as a
welder or caulker/burner.
In a follow-up study conducted 7 years later, Chinn et
al. (Ref. 48) reassessed pulmonary function and respiratory symptoms in the same cohort. Deaths (53/609) were
related to age, lung function, and smoking, but not to
working as a welder or caulker/burner. Of the 488 survivors who participated in the follow-up study, 425 had
retired.
Chronic bronchitis developed in 77 men during the 7
year follow-up period. This condition was significantly
related to smoking but not to trade as a welder or
caulker/burner. Breathlessness on exertion was associated with trade as a welder and with a history of pneumonia. The welders and caulker/burners were combined for
the analysis of spirometric tests, since their demographic
characteristics and lung function had not differed significantly in the earlier study. After adjustment for age, the
annual decline in lung function, as measured by FEV1,
FVC, and other spirometric parameters in the 488 participants, was greater among smokers and those who
had ever been welders or caulker/burners than among
controls.
The average decline in FEV1 for a nonsmoking nonwelder aged 50 was 16.2 mL/year. For a welder, the decline increased to 32.6 mL/year, and, for a welder who
smoked, the average decline in FEV1 was 50.3 mL/year,
three times that for a nonsmoking worker in other shipyard occupations. The annual decline in FEV1 was also
greater in men with respiratory symptoms (bronchitis,
wheezing, or breathlessness) than in those who were asymptomatic. In a sub-sample of 124 redundant workers,
there was also a significant interaction between the spirometric results (FEV1 and FEV1/FVC) and atopy (positive
response to skin tests with common allergens) or elevated serum immunoglobulin E (IgE) levels. Chinn et al.
(Ref. 48) concluded that exposure to welding fumes interacted with smoking and an atopic constitution to
produce respiratory impairment.
Bogadi-Sare (Ref. 31) examined lung function and the
incidence of respiratory symptoms among 47 stainless
steel welders and 59 nonwelders employed in machining
and polishing stainless steel. Controls were 80 workers

considered to be unexposed to irritant or allergenic pollutants in the workplace. The incidence of chronic bronchitis did not differ between stainless steel workers and
controls, but complaints of dyspnea (shortness of breath)
and choking were significantly more frequent among the
exposed workers. The stainless steel workers showed a
workplace-related, statistically significant decrease in
FEV1, FEV1/FVC, and other ventilatory parameters indicative of bronchial obstruction. Smoking did not contribute to differences between stainless steel workers and
controls. Contrary to the findings in other studies, there
were few significant differences in the spirometric parameters between similarly exposed smokers and nonsmokers. Bogadi-Sare stated that there was no difference
in lung function between the welders and the other stainless steel workers, but data to this effect were not presented. It was concluded that stainless steel dust is an
important cause of respiratory obstructive disorder in
industrial workers.
Kilburn and Warshaw (Ref. 111) examined the effects
of welding on pulmonary function in 226 male electric
arc welders in the midwestern United States. The subjects had welded for an average of 21.3 years. They had
no shipyard exposure and showed no radiographic signs
of asbestosis. Information concerning occupational exposures and respiratory symptoms was obtained by questionnaire. Nearly 20% had chronic bronchitis, and 11.3%
had a history of asthma; the incidence of these conditions
was the same in smoking and nonsmoking welders.
The performance of this group in lung function tests
was compared with a historical control group representative of the population of Michigan men. After adjusting
for height, age, and years of smoking, small but statistically significant reductions in lung function correlated
with years of welding experience. The only significant
difference observed between smoking and nonsmoking
welders was a slightly larger decrement in the FEF75-85 in
smokers. Using regression coefficients derived from data
collected in this study, the authors calculated that 40 years
of welding would reduce the FEV1 to 93.3%, the midflow parameter FEF 25-75 to 77.6%, and terminal flow
FEF75-85 to 62% of age-adjusted normal values in nonsmokers. In addition to declines in these parameters, there
would also be a reduction in FVC in welders who smoked.
The investigators concluded that chronic exposure to
arc welding gases and fumes causes airway obstruction
regardless of smoking or asbestos exposure. The magnitude of the decrements in lung capacity was smaller than
those attributed to asbestos exposure in a parallel study
of boilermakers with shipyard exposure conducted by the
same investigators (Ref. 110).
In a third study, Kilburn and Warshaw (Ref. 112) investigated whether changes in lung function occur during
a workshift in workers exposed to stainless steel welding
fumes. Pulmonary function and respiratory symptoms

31

were determined in 90 male welders employed for an average of 11 years at a fabricating shop making nuclear reactor vessels. Most of the welding in this shop was done
with GTAW using argon as the shielding gas. Neither respirators nor local exhaust devices were used. Air sampling was not conducted, but on the basis of impairment
of visibility, fumes in the welding area were estimated to
be 1 to 3 mg/m3. Data on respiratory symptoms were obtained by questionnaire. The rate of chronic bronchitis
among smoking welders was 18%, which was four to
five times greater than the rate in the reference population of Michigan men who were current smokers. In nonsmoking welders, the rate of bronchitis was 20 times the
rate for nonsmokers in the comparison group. There was
a small but significant reduction in FVC, FEV 1 , and
FEF25-75 in smoking welders compared with the reference population while only the FEF25-75 was reduced in
nonsmoking welders. No changes were seen in mean diffusing capacities for carbon monoxide or alveolar volumes in either group.
Pulmonary function, measured in 31 of the workers,
did not change across a work shift. Included in this phase
of the study were seven stainless steel welders, 14 blackplate steel welders, and ten nonwelders who worked in
the welding area and had only indirect exposure to welding fumes. Pulmonary function, and chromium and
nickel concentrations in blood and urine, were determined before and after the workshift, at the start of the
week. Serum and urine nickel and chromium levels were
higher in the study participants than in the reference population. Blackplate steel welders had no change in urine
or serum chromium levels during the workday. Serum
chromium levels increased by 66% during the workshift
in the seven stainless steel welders. Thus, changes in
chromium absorption were not accompanied by changes
in pulmonary function during an 8-hour workshift.
Lukac et al. (Ref. 128) assessed functional changes in
the lungs of 101 men who welded steel bridge components in an enclosed factory room in Czechoslovakia.
Lung disorders were found in 32 of 64 smoking welders
and in 15 of 37 nonsmoking welders. There were no nonwelding controls in this study. The severity of the lung
damage was related to duration of exposure to welding
fumes and to tobacco smoking. In smokers with 10 or
fewer years welding exposure, peripheral airways disorder was seen in 6 of 26 welders while another six had
obstructive lung disease. Among welders with over 20
years experience, obstruction of the central airways, indicated by decreases in FEV1 and FEV1/FVC, was seen
in 11 of 21 smokers and 4 of 14 nonsmokers. Dust concentrations were measured in 20 work areas during one
shift in this factory. The GDR allowable concentration of
10 mg/m3 for total dust was exceeded in five of the work
areas. The levels of iron, manganese and chromium,

measured during CO 2 -shielded welding, were within

the GDR allowable limits. Two out of three measurements of fluoride concentration exceeded the GDR limit
of 1 mg/m3.
Melbostad and Ruud (Ref. 136) studied the relationship between lung function, respiratory symptoms, and
exposure duration in 164 welder/plate workers and 134
machine workers in a mechanical production plant. Respiratory symptoms, as determined by questionnaire, were
significantly more prevalent in welders than in machinists. Multiple linear regression analysis showed a significant reduction in FEV1/FVC with years of smoking and
grinding but not with years of welding.
Rossignol et al. (Ref. 176) measured FEV1 and FVC
in 298 male welders and burners working in 31 metal
manufacturing plants. FEV1/FVC was the only measure
of lung function significantly related to number of years
working (seniority) as a welder/burner. FEV1/FVC increased 1.8% for each 10 years of seniority. FEV1/FVC
increased by 2.7% for 10 years seniority in workers
older than the median age in the study (38.3 years), while
lung function did not change with seniority in younger
workers. The findings of this study are contrary to other
published reports.
Kleiner et al. (Ref. 115) screened 2455 Russian workers engaged in machine manufacture for bronchitis. Of
the 1794 workers exposed to metal dust or fumes, 389
were electric arc welders. The controls were 661 mechanics with little dust exposure. Chronic bronchitis was
significantly more prevalent in the exposed group
(24.3%) than in controls (13.6%). Compared with the
mechanics, welders developed chronic bronchitis at an
earlier age and after briefer employment in their trade.
Twenty-nine percent of welders who smoked developed
this condition, compared with 3% of those who did not
smoke.
The prevalence of chronic pulmonary disease was assessed in 546 members of a Michigan boilermakers
union by Demers et al. (Ref. 53). Eighty-two percent of
the study participants were welders. Because of the nature of their work, boilermakers have historically been
exposed to asbestos. Since exposure levels were not
available, years in the trade was taken as an indicator of
overall asbestos exposure. Radiographic changes (opacities and pleural thickening), respiratory symptoms (rales,
chronic cough, wheeze, chronic phlegm, and dyspnea)
and pulmonary function deficits (FVC, FEV 1 , and
FEV1/FVC) were statistically more prevalent among the
boilermakers with increasing years in the trade. However, a statistical analysis, which included smoking status
and years in the trade, did not show a positive association
between welding and interstitial fibrosis or decrements in
FVC or FEV1.

32

Lubianova et al. (Ref. 127) reported the results of

medical surveys of welders in Kiev. They found that
35% of the welders had signs of chronic bronchitis. The
incidence increased with increased welding exposure,
age, and smoking.
Epidemiologic studies of occupational environments
are frequently hampered by the healthy worker effect,
i.e., workers who become ill from occupational exposures or other causes are more likely to quit the workplace or retire early, leaving behind a population of
workers that is healthier than persons of the same age in
the general population. To eliminate the influence of the
healthy worker effect on data concerning the contributions of occupational exposure to obstructive lung disease, Bakke et al. (Ref. 11) conducted a study of a
random sample of 4992 persons from a total population
of 267,304 persons in Hordeland county, Norway. The
objective of that study was to obtain estimates of the
prevalence of obstructive lung disease and occupational
exposure to asbestos or quartz.
A second phase of this study focused on the 1512 subjects who lived in Bergen and 11 nearby municipalities
(Ref. 10). A total of 1275 persons, including 84 welders,
participated in this phase of the study. Each of the participants was examined for chronic cough, phlegm when
coughing, and breathlessness or wheezing, which, when
accompanied by a ratio of FEV1/FVC less than 0.7, was
diagnosed as chronic obstructive lung disease. Questionnaires were used to obtain data on smoking habits and
past or present exposure to asbestos, quartz, wood dust,
chromium, nickel, or platinum fumes, aluminum production and processing, and welding or soldering. Fiftyseven percent of the men and 2% of the women had experienced occupational exposure to one or more of these
agents. Welding, along with exposure to quartz, metal
fumes, and aluminum production and processing, was
significantly associated with chronic obstructive lung
disease and asthma [odds ratio (OR) for welding = 2.2].
The investigators cautioned that the data should be interpreted carefully because of the small number of subjects
with reduced pulmonary function and because of the limited information obtained concerning occupational
exposures.
9.2 Case Reports. Lee et al. (Ref. 122) examined the 35
cases of occupational asthma reported to the Ministry of
Labor in Singapore between 1983 and 1990. Three of
these cases were related to exposure to welding fumes (2
SMAW and 1 spot welding). The prevalence of smoking
and the incidence of atopy did not differ significantly
from controls. All patients were trained to use a peak
flow meter and to record results. Plotting the PEFR every
3 hours during waking hours, on and off the job, for a period of 3 weeks, permitted evaluation of changes in the
mean and diurnal variations. Improvement of the PEFRs

during periods away from work, and deterioration of

PEFRs while in the work environment, were taken as evidence of work-related airway obstruction.
One case of occupational asthma, involving a female
welder who joined metal wires into a mesh for concrete
reinforcement by resistance spot welding, was reported
in detail (Ref. 121). This welder worked in a well-ventilated area; lubricating oil was the only chemical used in
the vicinity. She welded mild steel with copper electrodes containing 0.4% chromium. Nickel, chromium,
cobalt, and cadmium were not detected by environmental
monitoring during the spot welding process. Clinical examination showed allergy to house dust and nonspecific
bronchial hyperreactivity. Patch skin testing did not
show reactivity to specific metals. Peak expiratory flow
was measured at home and at work for about one month.
The PEFR improved during days spent at home and deteriorated during days spent at work. When she was transferred to a nonwelding job in the same plant, there was
no difference in peak expiratory flow between periods at
home and at work.
A case of siderosis (iron deposits in the lung) in a man
who had welded for 30 years was described by Lasfargues et al. (Ref. 120). While reticulo-micronodules
were seen in chest x-rays, pulmonary function was normal, and there were no clinical signs of disease. Numerous hemosiderin-laden macrophages were noted in
bronchoalveolar lavage (BAL) fluid, and analyses of biopsy tissue revealed patchy interstitial fibrosis. The investigators suggested that siderosis is not as benign as
previously thought and concluded that this welder may
have developed fibrosis as a result of siderosis, even in
the absence of exposure to crystalline silica. Because
many materials can cause pulmonary fibrosis, and pulmonary iron deposits are generally considered to be
non-fibrogenic, it is important to rule out all suspect exposures or the presence of nonferrous fibrous or fibrogenic materials in the lungs before drawing such a
conclusion.
A man who worked as a plasma arc welder was diagnosed with chronic bronchitis 12 years after he had
started welding. He continued welding for 3 more years
and then was placed in a series of nonwelding jobs, during which time the severity of his bronchitis continued to
increase. He died of heart failure at age 57, seven years
after quitting welding. A survey of the workplace atmosphere had been performed in 1974, in the midst of his
welding experience. Aluminum trioxide, iron oxide, and
manganese dioxide were well in excess of the GDR industrial standards at the time, but copper and Cr(VI)
were below the GDR standard. On that occasion, his
workplace was given an industrial hygiene rating indicative of a condition in which harmful effects or loss of
productivity were almost certain to occur (Ref. 217).

33

Autopsy revealed mixed dust pneumoconioses, chronic

tracheobronchitis, and massive emphysema. Histopathologic examination showed darkly pigmented deposits, particularly in the perivascular area, and free-floating coarse
granular material in the alveoli. Chemical analysis of lung
tissue showed the following concentrations of metals per
gram of tissue:
Iron

12,300 g/g

Nickel

600 g/g

Aluminum

10,400 g/g

Manganese

90 g/g

Chromium

980 g/g

Copper

30 g/g

According to Wagner et al. (Ref. 217), these values were

all at least ten-fold greater than those found in the general population.
A routine chest x-ray revealed small, bilateral nodular
shadows and a cavity in the left apical area of the lungs
of a 43-year old Japanese welder examined by Kishimoto
et al. (Ref. 113). After a year, during which he was
treated with antibacterial agents without apparent effect,
a lung biopsy revealed a tuberculous lesion, pneumonoconiosis, and asbestos bodies in the left apical cavity. A
mycobacterial infection (M. kansasii) was also found in
the cavity. The pneumonoconiosis and asbestosis in this
welders lungs were considered by the authors to be an
ideal culture medium for this infection.

10. Cancer
The International Agency for Research on Cancer
(IARC) evaluated epidemiologic studies reported
through 1989 which assessed the incidence of cancer in
welders (Ref. 100). IARC concluded that [t]here is limited evidence in humans for the carcinogenicity of welding fumes and gases. This evaluation indicates that a
positive association has been observed between exposure to [welding fumes] and cancer for which a causal interpretation is considered...to be credible, but chance,
bias, or confounding could not be ruled out with reasonable confidence. The evidence for carcinogenicity in
experimental animals was judged inadequate (...studies
cannot be interpreted as showing either the presence or
absence of a carcinogenic effect...), and the overall
evaluation of IARC was that [w]elding fumes are possibly carcinogenic to humans.
The agents in welding exposures that may be responsible for the cancer risk are not known. Stainless steel
welding emissions contain hexavalent chromium, which
is recognized by IARC and the National Toxicology Program as being carcinogenic in humans (Ref. 98). In addition, some forms of nickel are carcinogenic, but it is not

known if the form of nickel present in stainless steel

fumes is carcinogenic (Refs. 99, 133, 150, and 208).
Ironically, stainless steel welders do not appear to be at a
greater risk for lung cancer than mild steel welders.
In the various industrial settings in which they work,
or have worked in the past, welders may have been exposed to other substances which may act alone or in
combination with the components of the welding fumes
to increase the risk for cancer. For example, Tossavainen
(Ref. 202) pointed out that welders in foundries may
have substantial exposures to materials such as silica and
polycyclic aromatic hydrocarbons which could increase
their risk for lung cancer. These, and other factors unrelated to the work environment, such as smoking or
avocational exposures to substances which may act as
carcinogens, co-carcinogens, or tumor promoters, contribute to the health status of workers and are known to
confound epidemiologic studies of the cancer risk in
welders.
10.1 Lung Cancer. The cohort and case-control studies
considered by IARC (Ref. 100) in the evaluation of the
association between exposure to welding fumes and lung
cancer are listed in Tables 1 and 2. The cohort studies
compared mortality from cancer in welders with cancer
mortality, adjusted for age and sex, in the general population (the standardized mortality ratio or SMR) or with a
specific control population (the proportional mortality
ratio or PMR). In some instances, the incidence of cancer
from cancer registries was used (standardized incidence
ratio or SIR). The cohort studies considered by IARC are
shown in Table 1. [Where a confidence interval (CI) is
given, the SMR, PMR, or SIR is considered statistically
significant if the lower confidence limit exceeds 100].
While most of the large cohort studies conducted in Europe and the United States showed a greater incidence of
lung cancer among welders than among the control population, only one of these was significant (Ref. 186). In
addition to the published studies, IARC relied upon an
analysis of data combined from a number of epidemiologic studies. This analysis, published in 1990 by Simonato et al. (Ref. 184), showed a significant relationship
between welding and lung cancer mortality.
The twelve case-control studies examined by IARC
are shown in Table 2. Of these, two detected no excess
risk of lung cancer among welders or those exposed to
welding fumes (Refs. 29 and 30); four showed a moderate excess (Refs. 26, 114, 171, and 179), one of which
was significant (Ref. 171); and the remaining six showed
greater than a 100% excess risk (Refs. 39, 40, 76, 123,
174, and 183). The excess risk was statistically significant in four of the latter studies (Refs. 39, 76, 123, and
183). These four studies, along with the positive association of lung cancer with exposure to welding fumes seen
in the majority of the cohort studies and in the data

34

Table 1
Lung Cancer in Welders, Cohort Studies
No. of
Cases
Observed

SMR,
PMR or
SIR

95% CI

49

105

78139

125
212

34320
58542

Autogenous welders; two sets of standard rates

used (male population of Genoa and male staff of
hospital)

160
254

33466
52743

Electric welders; two sets of standard rates used

Polednak (1981) (USA)

(Ref. 164)

17
7
10

150
124
175

87240
50255
84322

All welders
Welders exposed to nickel compounds
Other welders

Beaumont and Weiss (1981) (USA)

(Ref. 23)

50

132

98174

McMillan and Pethybridge (1983)

(UK) (Ref. 134)

104

34243

Fletcher and Ades (1984) (UK)

(Ref. 65)

146

62288

26

113

80157

Shipyard welders; SMR, 191 compared with

general population of England and Wales (one
mesothelioma)

95

35207

Stainless-steel welders; expected numbers based

on national mortality statistics

170

70400

Cohort of machinists used as controls (two

mesotheliomas) (PMR)

Sjogren and Carstensen (1986)

(Sweden) (Ref. 186)

193

142

123163

Sjogren et al. (Sweden) (1987)

(Ref. 188)

249

80581

Stainless-steel welders

27
14

115
142

76167
77237

Welders in shipyards (SIR)

Welders in machine shops (SIR)

Reference (Country)
Dunn and Weir (1968) (USA)
(Ref. 56)
Puntoni et al. (1979) (Italy)
(Ref. 167)

Newhouse et al. (1985) (UK)

(Ref. 145)

Becker et al. (1985) (FRG)

(Ref. 24)

Tola et al. (1988) (Finland)

(Ref. 199)
Data from IARC, Ref. 100

Comments

PMR for respiratory cancer (three mesotheliomas)

Unadjusted SMR (four mesotheliomas)

35

Table 2
Lung Cancer in Welders, Case-Control Studies
No. of
Cases
Exposed

RR

95% CI

Breslow et al. (1954) (USA)

(Ref. 39)

14

7.2

1.944.3

Blot et al. (1978) (USA)

(Ref. 29)

11

0.7

Blot et al. (1980) (USA)

Ref. 30)

11

0.9

0.42.3

Rinsky et al. (1988) (USA)

(Ref. 171)

41
236

1.1
1.5

0.81.7
1.21.8

Probable welding exposure

Potential welding exposure

Buiatti et al. (1985) (Italy)

(Ref. 40)

2.8

0.98.5

Adjusted for smoking

Silverstein et al. (1987) (USA)

(Ref. 183)

13.2

1.1154.9

Kjuuss et al. (1986) (Norway)

(Ref. 114)

28
16

1.9
3.3

0.93.7
1.29.3

All welders; adjusted for smoking

Stainless-steel welders; adjusted for smoking

Gerin et al. (1986) (Canada)

(Ref. 76)

12
10

2.4
3.3

1.05.4
1.29.2

All welders
Welders exposed to nickel

Schoenberg et al. (1987) (USA)

(Ref. 179)

38

1.2

0.81.9

Welders or flame cutters; adjusted for smoking

Lerchen et al. (1987) (USA)

(Ref. 123)

19
6
13

3.2
2.2
3.8

1.47.4
0.59.1
1.410.7

All welders
Welders employed in shipyards
Welders not employed in shipyards

Benhamou et al. (1988) (France)

(Ref. 26)

18

1.4

0.792.9

Adjusted for smoking

Ronco et al. (1988) (Italy)

(Ref. 174)

2.9

0.879.8

Adjusted for smoking

Reference (Country)

Data from IARC, Ref. 100

Comments
RR 7.7 adjusted for smoking

36

assembled by Simonato et al. (Ref. 184), served as the

basis for the conclusion by IARC that there is limited evidence for the carcinogenicity of welding emissions in
humans.
In 1985, a group of European investigators agreed to
pool data from planned, or ongoing, cancer mortality
studies in order to increase the statistical power of their
analyses. The results were published by Simonato et al.
(Ref. 184; see also the review by Moulin et al., Ref. 141).
Twenty-one case-control and 27 cohort studies were included. The combined study population consisted of
11,092 male welders from 135 companies in 9 countries.
The welders were categorized as mild steel only,
stainless steel ever, and predominantly stainless
steel. Shipyard welders were placed in a separate category because of the probability of confounding asbestos
exposure. SMRs were calculated using mortality data
from the country in which each participating company
was located.
The overall mortality of the welders was less than expected as compared with the reference populations, consistent with the healthy worker effect. Of the major
causes of death, only those from malignant neoplasms
were higher than expected, based on comparison with
control populations. This excess (SMR = 113, CI = 100126) was of borderline significance, and was due primarily to a significant excess of deaths from lung cancer
(116 observed vs 86.81 expected; SMR = 134, CI = 110
to 160). The risk of death from lung cancer was higher in
mild steel only welders (SMR = 178) than in stainless
steel ever welders (SMR = 128). Deaths from lymphosarcoma (SMR = 171) and from cancers of the bladder
(SMR = 191), kidney (SMR = 139), and larynx (SMR =
148) were also higher than expected, but only the excess
bladder cancer was statistically significant.
Analysis of the data for the entire cohort of welders
revealed no tendency for the risk of lung cancer, or any
other type of cancer, to increase with time since first exposure. However, when the cohort was broken into subgroups, the lung cancer incidence tended to increase with
time since first exposure in all groups except shipyard
welders. This trend was only statistically significant
(p<0.05) for the predominantly stainless steel group.
The excess mortality from lung cancer was unrelated to
duration of employment, either in the entire cohort of
welders or in any of the subgroups. Estimated cumulative doses of total fume, total chromium, or hexavalent
chromium were not significantly associated with mortality from lung cancer in any of the subgroups.
The presence of five cases of mesothelioma in the cohort of welders indicated that exposure to asbestos could
be a confounding factor. The mesothelioma deaths were
distributed among all of the occupational subgroups,
suggesting that at least some welders in each category

had been exposed to asbestos. In addition, the effects of

tobacco smoking could not be ruled out, but Simonato et
al. (Ref. 184) maintained that smoking habits were unlikely to have been the reason for differences in lung cancer rates between welders and the reference populations.
Four epidemiologic studies of the lung cancer risk in
welders were published in 1990 and 1991. Becker et al.
(Ref. 25) conducted a follow-up to a study originally reported in 1985 (Ref. 24). The original study showed a
significant increase in cancer mortality among 1221
welders exposed to fumes containing chromium and
nickel as compared with 1694 machinists who worked in
the same plants and with the general population in the
Federal Republic of Germany. (This work was included
in the cohort considered by Simonato et al., Ref. 184). In
the follow-up study conducted 6 years later, the number
of deaths due to malignant neoplasms among welders
was still slightly greater than expected in the general
population, but this result was no longer significant
(SMR = 109, CI = 82-144). The earlier study had found a
significant excess of deaths from all malignant neoplasms among welders. The relative risk (RR) was 2.4
compared with machinists in the same plants. In the follow-up, the RR was reduced to 1.6, but the excess was
still significant. The RR remained unchanged after adjustment for smoking. Both welders and machinists had
slight, but nonsignificant, excesses of lung cancer when
compared with the general population. Cancer mortality
was not significantly associated with years since first exposure for any group of welders. The increased risk of
lung cancer found in this study was smaller than that
seen in the large international study (Ref. 184).
Steenland et al. (Ref. 192) conducted a mortality
study of 4,459 mild steel welders who had worked in
three midwestern United States plants from the mid
1950s through 1988 and had had no occupational exposure to asbestos or to stainless steel welding fumes. The
mortality of these welders was compared with the United
States population and with a group of 4,286 nonwelders
in the same plants. The primary welding technique in the
plants was SMAW until the mid-1960s and GMAW
thereafter. Personal monitoring data, measured between
1974 and 1987, were available for the three plants.
Welders were exposed to average concentrations of 6 to
7 mg/m3 total particulate and 3 to 4 mg/m3 iron oxide.
Nonwelders working near welding areas were exposed to
average concentrations of 0.5 mg/m3 total particulate and
0.16 mg/m3 iron oxide; these exposures were negligible
compared with those received by the welders. The lung
cancer rates for welders and nonwelders were not significantly different and were slightly, but not significantly,
higher than those of the general population (welders
SMR = 107; nonwelders SMR = 117). The lung cancer

37

risk among the welders did not correlate with duration of

exposure nor with time since first exposure.
Sankila et al. (Ref. 178) studied the relative lung cancer risk of different occupations in Finland by linking
data from the Finnish Cancer Registry (1971 to 1980) for
male lung cancer patients aged 25 through 64 years to
occupational information obtained from the 1970 Finnish
census. There were 57 cases of lung cancer among welders which corresponded to a statistically significant SIR
for lung cancer of 150. When the cancer incidence was
calculated for specific histological types of lung cancers,
welders had a statistically significant elevated incidence
of epidermoid carcinoma of the lung (RR = 1.97).
Gallagher et al. (Ref. 69) evaluated cancer mortality
in skilled manual metalworkers in British Columbia who
died between 1950 and 1984. The 125 deaths from lung
cancer among metal workers corresponded to a significantly elevated risk compared with other skilled manual
workers (PMR = 129, p<0.001). Significantly greater
PMRs were also found for Hodgkins disease (10 deaths,
PMR = 226, p<0.05), and all cancers (358 deaths, PMR
= 124, p<0.0001). Calculating PMRs within a single
class of workers allowed some control over smoking
rates, which were assumed to be strongly related to social class. The authors cautioned that the lack of precision in identification of cause of death and occupation is
a drawback in proportional mortality studies.
10.2 Nasal and Laryngeal Cancer. Dietz et al. (Ref. 54)
reviewed the recent literature concerned with carcinomas
of the head and neck region and concluded that the main
risk factors are chronic consumption of alcohol and use
of tobacco. Occupational exposures and dietary habits
were also associated with elevated cancer in this region.
Two case-control studies of cancer in the nasopharyngeal
region were reviewed by IARC (Ref. 100). In the first,
welders in general, and stainless steel welders in particular, showed increased rates of laryngeal cancer (Ref.
154). In the second case-control study (Refs. 85 and 86),
an increased incidence of nasal cancer was associated
with welding, flame-cutting and soldering.
A recent German study (Ref. 2) examined risk factors
for laryngeal cancer in 100 male patients hospitalized for
this disease compared with 100 male patients with unrelated diseases. Daily consumption of alcohol and heavy
smoking were strongly related to laryngeal cancer. Welders and burners did not show an increased risk for laryngeal cancer. The authors recognized that the small
sample size limited the ability of this study to detect
effects.
Opposite results were seen in a study by Yu et al.
(Ref. 227), who interviewed 306 nasopharyngeal carcinoma patients and 306 neighborhood-, age-, and sexmatched controls from Guangzhou City, China. Exposure to combustion products (63 cases, 33 controls) was

identified among the risk factors for nasopharyngeal carcinomas (RR = 2.4, p = 0.001). In this exposure group,
15 cases and eight controls were exposed to welding
fumes, but the significance of these numbers was not
reported.
10.3 Urogenital Tract Cancer. Among the studies reviewed by IARC (Ref. 100), bladder cancer was found to
be significantly elevated in one case-control study (Ref.
94) and one cohort study (Ref. 186). None of the other
four studies cited by IARC (Ref. 100) showed a significantly elevated risk of cancer of the urinary tract in welders. However, bladder cancer was significantly elevated
among welders (SMR = 191, CI = 107-315) in the multinational European study described by Simonato et al.
(Ref. 184). Kidney cancer was also elevated (SMR =
139, CI = 72-243), but this was not significant. The incidence of cancer at these sites was not related to duration
of exposure nor to time since first exposure. An increased incidence of prostatic cancer was seen among
shipyard and mild steel workers, but was not exposurerelated.
10.4 Cancer in Children of Welders. Bunin et al. (Ref.
43) reviewed studies of the association of specific types
of childhood tumors with parental employment in the
metal industry. These included Wilms tumor (a childhood kidney tumor), brain tumor, and acute nonlymphocytic leukemia. Wilms tumor has been linked with
paternal exposures to lead and hydrocarbons, and paternal welding exposures are also suspect. IARC (Ref. 100)
cited three case-control studies which investigated the
etiology of Wilms tumor. The first study (Ref. 108)
showed a nonsignificant excess of Wilms tumor among
children of welders; in the second one (Ref. 220), none
of the fathers of children with Wilms tumor were welders; the third study (Ref. 42) found a significantly increased odds ratio for Wilms tumor associated with
children whose fathers were in a job cluster exposed to
hydrocarbons, metals, or inorganic compounds. Within
this cluster, five cases and one control had fathers whose
occupation was welding.
Olshan et al. (Ref. 155) conducted a case-control
study of 200 children with Wilms tumor. The results in
this study were also nonsignificant, but, according to the
investigators, they were suggestive of an association between paternal occupation as a welder and development
of Wilms tumor. During the preconception period, fathers of six cases and one control were employed as
welders; fathers of five cases and one control were welders during the pre-natal period. Fathers of six cases and
two controls were employed as welders after the babies
were born.
Retinoblastoma, a malignant tumor in the retina of the
eye, occurs most frequently in young children. It occurs

38

in two forms, heritable and nonheritable. The heritable

form may be familial (traceable to previous occurrence
of the disease in ancestors) or nonfamilial. Epidemiologic studies have associated sporadic heritable (nonfamilial) retinoblastoma with preconception exposures
of the father, and laboratory animal studies have demonstrated that most of the germinal mutations associated
with this form of the disease are paternal in origin.
Nonheritable retinoblastoma may result from postconception exposure of the mother and fetus.
A matched case-control study, conducted in 34 North
American childrens hospitals (Ref. 43), included 201
cases of retinoblastoma. Of these, 19 were familial and
were excluded from the analysis, 67 were sporadic heritable, and 115 were nonheritable. Children with sporadic
heritable retinoblastoma were more likely than controls
to have fathers employed in the metal-working industry
(p = 0.02) or in the military (p = 0.04). More cases of
nonheritable retinoblastoma had fathers working in a
job cluster that included welders, machinists, and
paper-processing workers than did controls (p = 0.04).
The authors pointed out that the study contained so
many comparisons that any or all of the significant
findings could have occurred by chance. But, since the
findings for workers in the metal-manufacturing and
welder/machinist/paper-processing groups corroborated
earlier observations, they believed that they were less
likely to be chance associations.
10.5 Cancers Associated with Electromagnetic Fields.
The association between exposure to EMFs and development of cancer remains in dispute. Several epidemiologic studies have suggested an association between
exposure to EMFs and leukemia and brain cancer. Data
from in vitro studies showing that extremely low-frequency magnetic fields (ELF MFs) can have an effect
upon growth and development, lends support to the hypothesis that these types of magnetic fields may act as
tumor promoters (Ref. 106). On the other hand, according to Jauchem (Ref. 104), epidemiologic studies of exposed workers have produced limited and inconsistent
data, ...insufficient to establish a link between [EMFs]
and cancer.
Five epidemiologic studies of the relationship between exposure to EMFs and the incidence of leukemia
and cancer of the central nervous system reported in
1990 and 1991 included welders in the exposed population. Bastugi-Garin et al. (Ref. 16) studied 185 cases of
acute leukemia and 513 age- and sex-matched hospital
controls in France. A significantly increased risk of acute
leukemia was observed in workers in several occupations
with EMF exposures (e.g., electrical engineering technician, furnace worker, X-ray technician), but no association was found between acute leukemia and arc
welding. Another case-control study conducted in Poland

by Pachocki and Gajewski (Ref. 158) compared 895

adult leukemia patients with 910 controls. Information
concerning exposure to EMFs was obtained from all participants. No relationship was found between the risk of
leukemia and exposure to EMFs. Persons identified as
exposed to EMFs were capacitor-discharge welders, induction welders, short-wave diathermy workers, and heat
sealer operators.
Gallagher et al. (Ref. 68) examined mortality data
for 320,423 workers from British Columbia who died
between 1950 and 1984. There was an insignificant increase in mortality from brain and central nervous system cancer in the group of selected occupations
suspected of having high exposures to EMF. Although
welders were considered to have a very high magnetic
field exposure, they showed no elevated risk of death
from brain cancer. Gallagher speculated that the inclusion of oxyacetylene welders, with presumably little or
no occupational exposure to magnetic fields, could have
contaminated the study, reducing the chances for
observing an association between central nervous system cancer and exposure to EMFs during electric arc
welding.
Tornqvist et al. (Ref. 201) examined the incidence of
leukemia and brain tumors in Swedish workers with
potential exposures to EMFs. Cases of leukemia and
brain tumors, and information on their occupations, were
obtained from the Swedish Cancer Environment Registry
and the national census. Of the 133,687 men who had
worked in occupations with EMF exposures, 21,045
were classified as welders and flame cutters. Welders and
cutters showed no excess risk of leukemia; the ratio of
observed brain tumors to the number expected, on the
basis of the cumulative incidence in the reference population of almost 2 million Swedish working men, was of
borderline significance (46 cases, SMR = 130, CI = 100
to 170). In a sub-group of welders and cutters in iron and
steel works, the incidence of brain tumors was higher,
but still only of borderline significance (14 cases, SMR =
320, CI = 100-740).
In a Finnish study, Juutilainen et al. (Ref. 106) determined the relative incidence of leukemia, acute myeloid
leukemia, and central nervous system tumors among
workers presumed to be exposed to EMF and ELF MF.
The study population consisted of all Finnish males who
were classified as industrial workers in the 1970 Finnish
census. The cancer incidence data were taken from the
Finnish Cancer Registry for 1971 to 1980. Welders and
flame cutters had no excess incidence of either leukemia
(5 cases vs 5.0 expected) or central nervous system tumors (8 cases vs 9.7 expected).
10.6 Cancers Associated With Ultraviolet Radiation.
Ultraviolet radiation is produced during arc welding,

39

with intensities depending upon the process and welding

conditions. In addition to the UVB radiation (280 to
315 nm) associated with skin cancers from sun exposure,
welding arcs and plasma torches produce UVC radiation
(100 to 280 nm), a particularly damaging region of the
UV spectrum. UVC is also a component of solar radiation, but it is totally absorbed by ozone and water vapor
before reaching the earth. UVC has induced skin cancers
in experimental animals and cell transformations in
mammalian tissue cultures (Ref. 62). Cancers have also
been shown to be more prevalent in areas of the skin that
have been subjected to repeated trauma or erythema due
to excessive heating such as may be experienced by
welders (Ref. 62). Despite this evidence, skin cancer has
not been shown to be associated with welding in epidemiologic studies. Skin cancer was not reported in the
case-control and cohort studies reviewed by IARC (Ref.
100), but one case report of multiple epitheliomas in a
welder/cutter was mentioned (Ref. 175).
Uveal melanoma is the most common intraocular malignancy in adults but few risk factors have been identified.
Holly et al. (Ref. 93) conducted a case-control study in 11
states in the western United States to examine the possible
role of UV light in the etiology of uveal melanoma. Included in the study were 407 patients with uveal melanoma
and 870 randomly selected controls. Severe burns to the
eyes from welding, sunburn of the eye, and snow blindness
were grouped together, and exposure to any of these factors
carried a statistically significant relative risk of 7.17 for
uveal melanoma compared with the controls.

11. Metal Fume Fever

Metal fume fever, an acute illness of short duration
with flu-like symptoms, is caused by exposure to high
concentrations of metal oxide fumes. Metals which can
cause fume fever include zinc, copper, magnesium, aluminum, antimony, iron, nickel, cadmium, and tin. A typical case of metal fume fever was described by Heydon
and Kagan (Ref. 88). A farmer spent 5 hours cutting galvanized steel with a gas torch without wearing a mask.
Two hours later, he developed a cough and had difficulty
breathing. He felt chilled and had cramps in both legs.
His symptoms resolved within a day. Other symptoms
which are characteristic of metal fume fever, but were
not associated with this case, include fever, nausea, headache, thirst, polyuria, diarrhea, weakness, fatigue and
general malaise.
Because tolerance can develop with repeated exposure to metal fumes, metal fume fever frequently occurs
on Mondays, after a weekend break from exposure.
Hence, metal fume fever is sometimes referred to as
Monday fever. It is often assumed that this condition

has an immunological basis, but that is uncertain. The

pathogenesis of metal fume fever in humans was examined by Blanc et al. (Ref. 27) who studied cellular functions and biochemical responses to zinc welding fumes.
Blanc cited previous studies which demonstrated that
zinc oxide fumes, even in low concentrations, can cause
marked pulmonary inflammation (e.g., see Lam et al.,
Ref. 118), suggesting that the metal fume fever response
is initiated in the lung. This is substantiated by reports
that neither ingestion nor intravenous injection of zinc
will elicit fume fever in laboratory animals. Based on
this, Blanc hypothesized that polymorphonuclear leukocytes (PMNs), a type of inflammatory white blood cell,
responding to inhaled zinc oxide fumes, release cytokines
(mediators which elicit reactions in cells or tissue remote
from the cells where they originated). The cytokines, in
turn, produce the systemic, flu-like symptoms characteristic of metal fume fever.
To test this idea, Blanc (Ref. 27) examined the pulmonary response of volunteers to zinc oxide fumes. Using
standard eye and skin protection, but no respiratory protection, fourteen experienced welders welded galvanized
mild steel with standard cellulose-covered carbon steel
(cadmium-free) electrodes for 15 to 30 minutes in an enclosed chamber with controlled ventilation, temperature,
and humidity. Pulmonary fluids were obtained by BAL at
8 hours after welding from five of the participants (early
follow-up) and at 22 hours after welding from the remaining nine participants (late follow-up). The BAL fluid was
assayed for total and differential white blood cell counts
and for concentrations of two cytokines, interleukin-1 and
tumor necrosis factor (TNF), deemed likely to be present
following this type of pulmonary challenge.
Only minimal changes were found in pulmonary
function and airway reactivity at 1, 6 and 20 hours after
welding. The number of PMNs in the BAL fluid increased markedly with time after exposure and correlated
positively with zinc exposure at 8 hours. The numbers of
lymphocytes, PMNs and macrophages correlated positively with exposure at 22 hours. Little or no TNF or interleukin-1 was detected in the BAL fluid. The
investigators recognized that a number of cytokines are
released from inflammatory cells and that cytokines
other than the two assayed could be responsible for propagating metal fume fever. As the authors suggested, they
may have examined the wrong cytokines, and it is still
possible that cytokines, or a cytokine-like mechanism,
may mediate the syndrome of metal fume fever.

12. Effects on the Ear

Burns produced by flying sparks are a common ailment of welders. Burns of the ear are infrequent but can

40

be very damaging. Mertens et al. (Ref. 137) described 25

cases of injuries to the tympanic membrane that resulted
from welding sparks. In all cases, the burns occurred
while welding overhead or in a bent over position in a
confined space. Fisher and Gardiner (Ref. 64) described
five cases of such injury which ranged from minor burns
to the external ear to persistent tympanic membrane perforation and inner ear damage. According to these authors the welders visor type helmets offer little
protection to the ears when working in cramped conditions with the neck flexed or when working overhead.
Ear protection can be improved by a drape over the head,
by a lateral extension on the helmet, or by the use of nonflammable ear plugs.

13. Effects on the Eye and Vision

13.1 Cataracts. Chronic exposure to low levels of UV
light, especially UVB (320 to 290 nm), has been implicated in the formation of cataracts. Lasers emitting UV
radiation can induce permanent cataracts within 24 hours
of exposure and can damage the retina, iris and cornea.
Exposure to IR radiation from hot materials such as molten glass or steel can also cause of cataracts.
Animal experiments have shown that increases in the
lens temperature by just several degrees is sufficient to
produce cataracts. Okuno (Ref. 151) calculated the temperature distributions developed within a theoretical
model of the human eye exposed to IR radiation. His results suggest that IR cataracts induced in the workplace
result from absorption of IR by the cornea followed by
heat conduction to the lens. From the relationship between the incident irradiance and the lens temperature,
the threshold IR irradiances for cataract formation were
determined to be in the range of 163 to 178 mW/cm2 for
exposures lasting longer than 5 minutes under normal
conditions. For workers who perform heavy work at high
ambient temperatures, these values may be reduced by
50%. Okuno suggested that these threshold data could be
used to establish workplace exposure limits for IR radiation exposure to the eye. To protect against cataract formation, he recommended that IR irradiance be limited to
80 mW/cm2 for exposures over 5 minutes.
In a survey of occupational injuries that occurred in a
work force of 160,000 persons over a period of 23 years
in the GDR, Hanke and Karsten (Ref. 83) identified four
welders with cataracts. The four welders were 29 to 56
years old at the time their cataracts were diagnosed and
had welded for 12 to 31 years. The IR exposures received by these men during their welding experience
were estimated from detailed exposure histories and reconstruction of their workplace routines. The average
number of hours per day spent welding was calculated

from measurements taken during 105 typical work shifts.

The intensity of the IR radiation they had received was
approximated by measuring the IR radiation produced by
the various welding processes they had used and calculating the energy that would have contacted their eyes,
taking into consideration the types of face shields they
had used and the length of time spent welding during a
shift. All four welders had routinely been exposed to
doses of IR radiation in excess of the maximum safe limits recommended by Okuno (Ref. 151). One of the four
welders had routinely been exposed, without eye protection, to radiation from preheated materials. This source
of IR radiation accounted for 37% of the calculated lifetime dose he received. Another of the welders received
radiation from metal parts that remained super-heated
after welding was completed. The welder did not wear
eye protection, and this radiation contributed 97% of the
total calculated IR radiation dose to the eyes. This man
was the youngest, and also had the briefest welding experience of the welders with cataracts; but all four developed cataracts at a younger age than is common in the
general population.
13.2 Photokeratitis. Shortwave UV light (270 to
290 nm) is absorbed by the outer layers of the eye and
can cause photokeratitis (arc eye, welders flash, or
photophthalmia). Photokeratitis, a marked inflammation
of the cornea, is the ocular condition most frequently encountered by welders. Symptoms appear between 1 and
24 hours, depending on the intensity of the exposure, and
include blurred vision, headache, photophobia (sensitivity to light), tearing, and conjunctivitis. These effects
normally last up to 2 days and have no sequelae.
While photokeratitis can be produced by unprotected
exposure of even a few seconds to a high-intensity UV
source such as arc welding, retinal damage usually results from exposures of longer duration. Power et al.
(Ref. 166) described the case of a 45-year-old welding
trainee who developed bilateral maculopathy (retinal injury) following unprotected exposure of less than two
minutes duration to a SMAW unit. The day after exposure, the patient experienced blind spots and distorted
vision but no ocular pain. While there was no photokeratitis, an edematous lesion was noted in the macula
(an area of the retina) of each eye. His visual acuity was
completely restored within 6 months after exposure. The
severity of the macular lesions was unusual for such a
brief exposure, as was the formation of retinal lesions in
the absence of photokeratitis. The investigators suggested a link between these observations and the drug
fluphenazine, which the patient had taken for 10 years to
treat his depression. They postulated that the drug
fluphenazine accumulates in the retinal pigment epithelium, where it can act as a photosensitizing agent, making the retina particularly susceptible to photochemical

41

damage. They cautioned that persons taking phenothiazine drugs may be at special risk of developing weldinginduced retinal damage.
Neki (Ref. 144) questioned the assumption that the
macular lesions were due to drug-related photosensitivity. Instead, he suggested that it is equally possible that
the drug concentrated in the cornea, making it more resistant to photochemical damage and allowing the formation of retinal lesions in the absence of corneal effects.
As Neki alludes to, these hypotheses are conjectural and
could be tested in experiments with laboratory animals.
13.3 Eye Injury. Occupational eye injuries comprised
2.5% of all injuries treated in a Hospital in Esbjerg, Denmark, during a 4 month period in 1987. Seventeen percent of the patients with eye injuries were welders (Ref.
9). Noting that ocular foreign bodies are an occupational
hazard for people working in metal processing industries, Banerjee (Ref. 13) performed a survey of patients
who came to the hospital in Walsall, England, with this
type of injury during a six-month period beginning in
August 1989. Of the 472 patients who appeared in the
emergency department with foreign body eye injuries,
164 were included in this study; 19 were welders. Most
of the patients surveyed were employed by the many
metal processing plants in the vicinity of Walsall. For
136 of the participants, the injury was sustained while
doing their usual jobs in the workplace. Sixty-five had
not been using the eye protection required by British
standards. Since the data were not broken down in terms
of occupation, it is not possible to discern whether the
welders were among those using eye protection, or even
if they were injured while welding.
A survey conducted by Ten Kate and Collins (Ref.
198) also related eye injury to the use of appropriate eye
protection, but the population studied by these investigators was restricted to welders. The welders selected for
this study did not necessarily seek medical care for their
injury, so less acutely serious eye injuries were recorded.
Welders were queried by two routes: 87 were interviewed directly by one of the investigators, and 41 responded to questionnaires sent by mail. Participants were
asked about the number of flashes received, or ocular
or systemic symptoms experienced during the week prior
to receiving the questionnaire. These data were related to
the type of welding conducted, use of eye protection, and
welding history. Forty percent of the welders reported
receiving at last one welding flash. Nine oxyacetylene
welders welded without eye protection during the previous week, but only one of them worked this way for
more than an hour. Unprotected exposure to nearby
welding was experienced by 64% of the welders. Eightyfive percent of the welders thought their eye protection
was in compliance with Australian standards; the remainder were unsure. Reports of eye symptoms (e.g.,

tired, sore, watery or itchy eyes) correlated well (p<.001)

with the number of flashes received. Blurred vision was
experienced by 27% of the welders, and back pain, headache, and neck pain were reported by about one-third of
the respondents.
Narda et al. (Ref. 142) studied the incidence of
chronic conjunctivitis over a 10-year period in arc welders working in a foundry in Italy. The study included 41
welders; 31 performed SMAW and 10 performed submerged arc welding. Airborne dust levels in the foundry
often exceeded the ACGIH TLV of 10 mg/m3 for total
dust. The frequency of chronic conjunctivitis was substantially higher among the welders than among controls
(RR = 4.25). The incidence of conjunctivitis was 49.8%
in the exposed group. Shielded metal arc welders had a
higher incidence (56.9%) than did submerged arc welders (32.4%). Ocular symptoms (eye burning, tearing,
photophobia) often occurred before the development of
conjunctivitis. The investigators concluded that findings
of a prevalence of ocular symptoms should lead to careful examination of the working environment.
13.4 Long-Term Effects. To explore the effects of longterm welding exposure on the cornea, Norn and Franck
(Ref. 149) examined the outer part of the eye in 217 male
welders from two factories in Denmark. The most striking finding was the presence of spheroid degeneration in
24% of the welders, compared with 4% in a control
group of nonwelders. This disorder is characterized by
small (30 40 m), globular, limpid, yellowish or brownish lesions on the exposed part of the conjunctiva or cornea. A significant, but much smaller, increase in the
incidence of pinguecula (yellow spots on the exposed
conjunctival bulbi) was also noted. Corneal cicatrices
(small scars on the cornea) were present in about half the
subjects. The incidence of these three types of lesions increased with increasing welding exposure. Pterygium, a
UV light-induced membranous growth on the outer eye,
was not related to welding in this study, as it was found
in only one welder and no controls.
13.5 Contact Lenses. The dangers of wearing contact
lenses while welding have long been debated. Rumors
that radiation from the welding arc could cause the lens
to fuse to the cornea, potentially causing blindness, stem
from an incident which occurred in Baltimore in 1967.
At that time, an arc welder at Bethlehem Steel Corp. was
exposed to an explosion at an electric switch box while
he was connecting a welding instrument to a 440 Volt
grid. The welder was wearing contact lenses and safety
glasses at the time of the incident. Following the incident, he suffered temporary injury to the corneal epithelium. The attending physicians determined that the injury
was unrelated to his exposure to the electric flash. Instead, they attributed the injury to failure to remove the

42

contact lenses for 17 hours following the incident. Reports of this incident soon became distorted, and the injury was described in American and international
journals and news media as a fusion of the lens to the
cornea caused by exposure to the welding arc. These and
other reports of blindness or vision loss resulting from
wearing contact lenses while welding have led to widespread company policies forbidding welders to wear contact lenses.
Several articles disputing such policies were published during the current report period (Refs. 77, 125,
168 and 222). Winkleman (Ref. 222) wrote in response
to two recent publications in the GDR which claimed
that recent research presented new evidence of health
risks associated with wearing contact lenses while welding. Giroux and Remba (Ref. 77) and Loriot and Tourte
(Ref. 125) reviewed the literature on this subject. They
described conclusions drawn by a study group sponsored
by the French National Institute of Research and Safety
and an extensive investigation conducted by the American Optometric Association. The latter investigation
showed that reports that a welding arc or electrical spark
caused fusion between the contact lens and the eye lack
scientific validity. The general conclusion of both groups
was that, when proper safety eyewear is worn over contact lenses, the use of contact lenses adds no further risk
to injury from exposure to the welding arc.

14. Effects on the Skin

Ultraviolet light, generated by electric arc welding,
can cause erythema (e.g., sunburn) on unprotected skin.
Ultraviolet radiation can be reflected by some surfaces
and reflected UV light, is still capable of causing
erythema or other forms of UV-induced injury. Hindsen
et al. (Ref. 92) described a case where a welder developed mild facial erythema even though he wore a welding helmet. Simple detective work showed that the
erythema developed after he started wearing a white textile hood in addition to the helmet. The erythema, which
was attributed to UV radiation reflected from the white
hood, could be prevented by a sunscreen or by the use of
black cloth.
Young et al. (Ref. 226) described a welder who had
recurrent facial dermatitis associated with welding. The
dermatitis disappeared when he was not welding and recurred within 1 day after returning to the job. Skin testing revealed no allergies related to work place exposures
but other tests showed an abnormal reaction to UVC and
UVB radiation and mildly abnormal reactions to UVA
radiation. A shield was used while welding so the investigators concluded the reaction was due to indirect exposure to UV radiation.

15. Effects on the Nervous System

Sjogren et al. (Ref. 189) conducted a study of neuropsychiatric symptoms among welders which focused on
effects of lead, manganese, and aluminum. Lead and
manganese have well established neurologic effects.
Chronic exposure to manganese can cause behavioral
changes, psychosis, and a Parkinson-like syndrome.
Acute exposure to lead has been associated with encephalopathy, and chronic exposure has been associated with
peripheral neuropathy, chronic encephalopathy, and
other neurobehavioral disturbances (Ref. 124). Studies in
kidney dialysis patients (see citations in Sjogren, Ref.
189) and in animals (Refs. 22, 116, and 195) have implicated aluminum as a neurotoxin. Aluminum is not well
absorbed, is usually cleared through the kidney, and is
not known to cause neurological disorders in healthy
individuals. However, in patients with reduced renal
function, it may accumulate and become a problem (Ref.
139).
The study cohort consisted of 65 welders who welded
primarily aluminum and 217 railroad track welders. One
third of the railroad track welders had welded leadpainted steel and high-alloy manganese, exposures less
common among the aluminum welders. Eighty-five percent of the aluminum welders and only 48% of the railroad track welders had welded for more than 10 years.
Information on exposures to welding fumes, disturbances of memory and concentration, affective changes,
and symptoms of vegetative nervous system dysfunction
was obtained by self-administered questionnaires. Subjects who welded aluminum, lead, or manganese for long
periods experienced significantly more neuropsychiatric
symptoms than welders exposed to chromium or nickel.
Depression and difficulty in concentrating were associated with exposure to aluminum fume. Short memory,
forgetfulness, and frequent headache were associated
with welding lead-painted steel. Experiencing painful
tingling sensations was associated with exposure to manganese fume. These results should not be considered
conclusive because (1) data on exposure to other neurotoxic agents (e.g., organic solvents, alcohol) were not
collected in this study and (2) the analyses were done
strictly by questionnaire. As the authors suggested, detailed psychometric studies should be performed on
welders exposed long-term to specific metals to verify
or refute these results.
The etiology of Parkinsons disease is not understood,
but some authorities believe it is linked to environmental
factors (Ref. 196). Chronic exposure to manganese can
cause a progressive disease, which in advanced stages
resembles Parkinsons disease, with muscle weakness,
muscle rigidity, tremors, and impaired gait. Wechsler
et al. (Ref. 218) reported the results of a pilot study

43

undertaken to evaluate a broad range of occupational and

environmental agents for potential associations with Parkinsons disease. The study compared 34 Parkinsons
disease patients (average age 68.4 years) with 22 neurology clinic patients (average age 58.9 years) with other
disorders. A self-administered questionnaire was used to
collect data concerning occupation, well-water use, pesticide use, metal exposures, medical history, smoking, alcohol consumption, and drug use.
Analysis by metal exposure in males indicated more
frequent exposures to aluminum and copper. Three of the
male Parkinsons disease patients and none of the controls had been employed as welders. The three welders
reported exposures to aluminum but not to other metals.
None of the Parkinsons disease patients reported exposure to manganese, mercury or nickel whereas one of the
controls had mercury exposure. Analyses of metal exposures were limited to males because too few of the females in the study reported such exposures. Thus, the
number of cases considered was extremely small (19
male Parkinsons disease cases vs 9 controls) and the results were not statistically meaningful.
Armon et al. (Ref. 7) conducted a case-control study
to evaluate risk factors for amyotrophic lateral sclerosis
(ALS), an uncommon degenerative disease of the motor
neurons. Occupational and recreational data were collected for 47 male patients and 47 corresponding patient
controls. Men with ALS had spent significantly more
time welding or soldering than controls (p<0.01). The
type of welding or soldering, and the materials worked
with, were not investigated. ALS patients had a greater
exposure to lead than did controls (p<0.05). Because the
latter results fit in well with data that had been previously
reported, the investigators concluded that their results
suggested an association between ALS and exposure to
lead vapor.

16. Effects on the Immune System

Ulrich et al. (Ref. 207) examined absorption of metals
and changes in serum proteins in welders who conducted
GMAW with a CO2 shield. Air concentrations of total
particulate in two welding areas were 5.4 mg/m3 and 6.3
mg/m3; 30% of the measurements were over 10 mg/m3.
The levels of Co, Cd, Cu, Cr, Mn, Ni, and Fe were measured in hair samples from 69 welders and 47 controls.
Cr, Mn, Ni, and Fe were significantly elevated (p<0.001)
in welders. There was a positive association between the
number of years spent welding and the level of Mn, Ni
and Fe. Serum levels of immunoglobulins IgG, IgA, and
IgM were measured as indicators of the activity of the
immune system and various serum proteins were measured as indicators of inflammation in 97 welders and 36

controls. Significant changes in humoral immunity were

seen in the welders; IgG increased, and IgM decreased
with the duration of welding. Likewise, the serum indicators of inflammation, complement and alpha-1-antitrypsin, changed significantly with the number of years
welding experience.

17. Effects on the Musculoskeletal

System
Welding often requires sustained effort in awkward
postures while handling heavy equipment, resulting in
static stress on the arms and shoulders. While this stress
occasionally results in traumatic injury, chronic disease
is more common. Torner et al. (Ref. 200) evaluated the
effects of welding on the musculoskeletal system by
comparing the frequencies of physical disorders in 58 licensed Swedish shielded metal arc and gas tungsten arc
welders and 33 office clerks from the same energy equipment production plant. Information concerning subjective symptoms was obtained by questionnaire. Objective
evaluation of musculoskeletal disorders was determined
by orthopedic examination. Figure 3 illustrates the frequency with which confirmed physical disorders and
subjective symptoms were detected in different parts of
the musculoskeletal system. Both physical abnormalities
and subjective symptoms were more prevalent among
the welders than among the clerks. The excessive shoulder level work performed by welders at this plant was reflected in the incidence of neck and shoulder symptoms;
76% of the welders compared with 36% of the clerks reported having experienced neck and/or shoulder symptoms during the previous year. This was confirmed by
clinical examinations which showed that the range of
motion in external rotation of the shoulders was significantly less among the welders than among the clerks. Almost half the welders had atrophied shoulder muscles,
the supra- or infraspinatus muscles being the most severely affected. The muscular atrophy correlated with a
decrease in 90o arm abduction strength.
Subjective knee symptoms and clinically confirmed
bursitis in the knee were significantly more prevalent
among welders than among controls. This was attributed
to the large amount of kneeling required of the welders.
Most of the welders (60%) considered their symptoms to
have been caused by heavy work for a prolonged time,
while only 12% attributed their musculoskeletal symptoms to a sudden traumatic incident at work. This was
in accord with the finding of Hyytiainen and Saarela
(Ref. 96) that only a small number of welders attributed
occupationally related lower back pain to sudden
movement.

44

Figure 3Frequency of Physical Symptoms Affecting Different Parts of the

Musculoskeletal System (Indicated by Solid Lines). The Proportion of Workers
Who Had Experienced Symptoms During the Previous 7 Days, or Who Had
Been Forced to Reduce Their Daily Activity as a Result of Physical Symptoms
During the Previous 12 Months, is Indicated. Torner et al., Ref. 200

Hyytiainen and Saarela (Ref. 96) surveyed risk factors

for accidental lower back injuries in 3500 persons employed in a shipyard and 700 workers employed in a
metal products manufacturing plant. Risk factors for accidental back injuries were identified by questionnaires,
observation, and insurance company accident reports.
Five of the eleven insurance company reports involving
lower back injuries in welders were attributed to lifting,
another five were due to stumbling, slipping, or falling,
and the remaining injury was caused by sudden movement. The major risk factors for accidents involving
lower back injuries identified by the welders in response
to a questionnaire were prolonged/awkward working
postures (44%) and incorrect heavy lifts (29%).
In a later study, Hyytiainen and Uutela (Ref. 97) explored the relationship between job stress and lower back
pain. Data were obtained by questionnaires sent to a
group of manual workers (182 welders, 197 plumbers

and 125 sheet metal workers) and 170 clerical workers

from a ventilation equipment factory in Finland. Ergonomic factors (e.g., prolonged sitting and standing, confined work space, and awkward work postures) had the
greatest influence on the development of lower back pain
among the manual workers. But nonphysical factors
(hurry, monotony, and an accelerated working pace)
were also related to lower back pain.
Jarvholm et al. (Ref. 103) used simultaneous measurement of intramuscular pressure and electromyography (tracings of the electrical changes which accompany
muscular activity) to study the effectiveness of an arm
support in reducing the load on the supraspinatus muscle
of the shoulder during simulated welding operations. Although both methods demonstrated that the arm support
achieved a significant reduction in the supraspinatus
muscle load during low- load assembly work and highload welding applications, the muscle pressure was still

45

high enough during the simulated welding operation to

reduce blood flow in the muscle. The authors concluded
that arm support greater than the 2.2 to 3.4 pounds
(force) used in this study would be needed to reduce intramuscular pressure sufficiently to lower the incidence
of shoulder pain and muscular injury.
Marciniak and Badowski (Ref. 132) reviewed 28,324
X-rays of the spine of job applicants and workers in a
Polish automobile factory. A high incidence of vertebral
epiphysitis (inflammation of vertebrae: 22.5%) and
scolioses (curvature of the spine: 10%) was noted among
the 17- to 21-year-old job applicants. Statistical analysis
showed that at age 40, static changes in the spine were
more strongly related to pre-existing vertebral epiphysitis than to the character of the work performed. However,
the highest incidence of scoliosis was found among
workers in jobs involving the greatest spinal loading, i.e.,
tinsmiths, pressers, and welders.

Carpal tunnel syndrome (CTS) results from nerve

damage caused by compression of the carpal tunnel
formed by the carpal bones of the wrist. This section of
the wrist is illustrated in Figure 4. CTS can result from
repetitive wrist movement tasks such as welding (Ref.
160) or from use of vibrating hand-held tools (Ref. 82).
The symptoms of CTS can range from numbness and
tingling in the hands, to wrist pain, and disability. Hagberg et al. (1991) conducted a study of 41 workers who
had required surgery to correct the effects of CTS. Welding was one of the four most common occupations
among the CTS patients. Panio (Ref. 160) stressed that
prevention of CTS is possible with the cooperation of
management, supervisors, and workers. He recommended a program that includes pre-employment screening to determine those who already have CTS symptoms;
education to make workers aware of causes and preventive measures; modification of job routine to reduce

Figure 4Drawing of Hand Showing Location of Carpal Tunnel and Position

of the Median Nerve Affected by Carpal Tunnel Syndrome. Panio, Ref. 160

46

unnatural angles of the wrist; special gloves or braces to

maintain correct wrist alignment; and tool modifications,
such as curved-handled welding guns.

18. Effects on the Kidney

18.1 Cadmium. Chronic exposure to low levels of cadmium fumes can cause kidney damage. Proteinuria, the
urinary excretion of low molecular weight (LMW) serum
proteins such as beta-2-microglobulin, is often the first
symptom of kidney damage and is generally thought to
result from damage to the kidney tubules. In 1984,
Chiesura, Trevisan, et al. (Ref. 47) reported the results of
studies of 16 persons from workshops where brazing was
performed with high-cadmium alloys. Beta-2-microglobulin and other proteins indicative of tubular dysfunction
were present in urine samples from some of the workers
who had no frank indications of kidney damage.
In a follow-up of that study, Trevisan and Maso (Ref.
204) re-examined six of the welders who had been
exposed to high cadmium levels before they ceased
welding in 1982. Excretion of LMW proteins typical of
tubular dysfunction, and serum and urinary cadmium,
were determined annually in each of the welders for the
first 5 years after they ceased welding. Cadmium levels
in urine and serum decreased steadily in all six subjects
during the 5-year follow-up period (Figure 5a and 5b).
Three of the subjects had shown no indications of LMW
proteinuria while welding, and their urinary protein levels
remained low during the 5-year follow-up period. The remaining three welders had shown LMW proteinuria
(beta-2-microglobinuria), and one of them also had developed total proteinuria, while still exposed to cadmium.
During the follow-up period, beta-2-microglobinuria
increased up to a maximum in two of these welders,
while in the third there was a temporary increase in beta2-microglobinuria which returned to normal levels after
5 years (Figure 5c). Total protein was only elevated in
the welder who had had this condition while exposed to
cadmium. These results indicate that tubular dysfunction
may be reversible in some, but not all, persons with this
cadmium-induced condition.
18.2 Chromium. While acute kidney disease can result
from massive exposure to chromium, chronic renal disease resulting from occupational or environmental exposure to chromium has not been reported. Based on
published literature, Wedeen and Qian (Ref. 219) developed an argument for the possibility that long-term exposure to chromium can induce chronic kidney disease.
Occasional cases of acute tubular necrosis following
massive absorption of chromate have been described in
man. In experimental animals given large doses of
hexavalent chromium compounds, chromate selectively

accumulates in the convoluted proximal tubule and

causes necrosis. Wedeen surmised that long-term exposure to low levels of chromium could also cause chromate accumulation in the kidney tubules of humans,
resulting eventually in kidney damage. Such damage
should be evidenced by the findings of LMW proteins in
urine. In this regard, studies of urinary proteins in chromate workers have yielded conflicting findings. Wedeen
cited one study in which LMW proteinuria was found in
chromium workers (Ref. 66). However, other studies
have not associated LMW proteinuria or other indicators
of renal function with long-term chromium exposure in
welders [Refs. 215 and 231). In addition, as Wedeen
pointed out, LMW proteinuria occurs after a variety of
physiologic stresses, is usually reversible, and cannot by
itself be considered evidence of chronic renal disease.
Nonetheless, Wedeen maintained that since high-level
chromate exposure can induce acute tubular necrosis,
and since one study showed LMW proteinuria in chromium workers, there is a strong need for large-scale,
prospective case-control epidemiologic studies to demonstrate whether or not delayed renal effects can result
from low-level, long-term exposure to chromium.

19. Fertility
Boshnakova and Karev (Ref. 37) examined the risk of
spontaneous abortions and stillbirths (reproductive failures) among families of welders. Data were gathered by
questionnaire from a study population consisting of 72
welders and 41 controls. Reproductive failures occurred
in families of 15.3% of the welders and 9.8% of the control group. Fetal deaths occurred earlier in the pregnancy
among welders families than among controls. Thus,
spontaneous abortions occurred significantly more frequently in the group of welders and still births occurred
significantly more frequently in the control group.
Bonde (Ref. 32) conducted a case-control study of the
association between welding exposure and delayed conception (subfertility), defined as no conception within at
least 2 years of unprotected intercourse. The study population consisted of 432 male welders and 240 nonwelding male metal workers and electricians from six plants
in Denmark. Data concerning reproductive experience,
occupational exposure, and medical history were obtained by self-administered postal questionnaires. The
response rate was 79% for welders and 83% for nonwelders. A significantly increased risk for delayed conception was observed among the welders.
In a continuation of this study, Bonde et al. examined
fertility in a Danish cohort of 3702 male metalworkers
(Ref. 35). The subjects were employed at 74 stainless
steel and five mild steel factories for at least 1 year

47

Figure 5A Cadmium Levels in Serum from Six Welders During the First 5 Years
After Exposure to Cadmium Ceased. Trevisan and Maso, Ref. 204

Figure 5B Cadmium Levels in Urine in Six Welders During the First 5 Years
After Exposure to Cadmium Ceased. Trevisan and Maso, Ref. 204

48

Figure 5CConcentrations of Beta 2-Microglobin in Urine of Six Welders During the

First 5 Years After Exposure to Cadmium Ceased. Trevisan and Maso, Ref. 204

between April 1964 and December 1984. Data on occupational exposures and history, welding method used,
and drinking and smoking habits were obtained by a selfadministered postal questionnaire in 1986. Birth rates
among welders were compared with those of men selected from the Danish Central Population Register.
Among subjects who had ever welded, the probability
of fathering a child during years when they were not
welding was significantly greater than that of metalworkers who had never welded (OR = 1.27; p<0.001).
Because of this, the investigators considered it inappropriate to compare fertility of ever welders with those who
had never welded. Instead, the analysis was restricted to
subjects who had ever welded, and fertility during years
at risk from welding was compared with fertility during
years with no welding exposure. Among persons who
had ever worked as welders, the probability of fathering
a child was slightly, but significantly, reduced during the
year following a year of welding exposure (OR = 0.89;
P = 0.005). The reduction in fertility was associated with
the welding of mild steel (OR = 0.86; P = 0.03) but not

with SMAW or GTAW of stainless steel (OR = 0.98).

The authors concluded that welding mild steel is associated with a slight, but significant decrease in male
fertility.
To determine whether these observations correlated
with changes in physiological parameters, Bonde (Ref.
33) compared semen quality and serum levels of sex hormones [follicle stimulating hormone (FSH) and luteinizing hormone] among 35 stainless steel (GTAW) welders,
46 mild steel (SMAW and GMAW) welders and 54 nonwelding metal workers. The study participants were from
six workplaces in Denmark. Mean exposure to fume
particulates was 1.3 mg/m3 for stainless steel welders,
and 3.2 and 4.7 mg/m3 for mild steel welders with low
and high exposures, respectively. These levels were
within Danish process-specific threshold limit values for
welding.
The sperm count, proportion of normal sperm forms,
degree of sperm motility, and the linear penetration rate
were significantly decreased in mild steel welders, while
there was no significant change in the sperm concentra-

49

tion of FSH. Follicle stimulating hormone concentrations

increased with increasing welding exposure and, with the
exception of sperm count, semen quality decreased. The
effects were less marked in stainless steel welders who
had significantly decreased semen volume, total sperm
count, proportion of motile sperm, and concentration of
serum testosterone.
To determine if the changes in semen quality were reversible, semen quality was examined in 19 of the mild
steel welders, 18 of the stainless steel welders and 16 of
the non-welding metalworkers before, and 3, 5, and 8
weeks after a 3-week, welding-free vacation period (Ref.
34). Three post-vacation collection time points were included to allow detection of changes that occurred at different stages in sperm development. No significant
improvement in any of the semen parameters was observed at any of the post-vacation time periods relative to
the pre-vacation values.
While Bondes studies suggest a reduction in fertility
and semen quality with welding, the changes observed
were slight, albeit significant. A positive dose-response
was not obtained when fertility was considered in terms
of years of welding exposure. An increase in effects with
increased years of exposure would have strengthened the
observations.
Never welders were excluded from the large cohort
study (Ref. 35) because their fertility rate was lower than
that of welders during periods of nonexposure. The rationale for excluding nonwelders from the study is not
clear, and it is possible that a direct comparison between
the overall fertility rates among welders (including periods of both welding and nonwelding) and nonwelders
would have shown no difference between the two
groups.
Bondes conclusion that welding reduces fertility was
based on differences in birthrates between periods of
welding and nonwelding in the same cohort. Alternatively, physical or psychological stresses encountered
during welding could have decreased interest in fathering
a child; the welders could have delayed conception, either consciously or subconsciously, until periods when
vocational stresses were reduced. Thus, differences in
the fertility of welders between periods of welding and
nonwelding may not actually reflect physiological
changes brought about by welding exposures. Perhaps a
more valid comparison would be between men who were
never welders and those who worked continuously as
welders during their entire family raising period.
Finally, while the investigation of semen quality supported the concept that welding is associated with decreased fertility, there were some inconsistencies
between the findings of the fertility and semen quality
studies. These are (1) a reduction in semen quality was
observed in stainless steel welders whose fertility rates

were not altered in the epidemiologic study and (2) fertility, but not semen quality, improved during a period of
nonwelding. As Bonde pointed out, the observed
changes were small and thus subject to confounding factors which are impossible to control with human populations. Because the effects are small, controlled studies in
laboratory animals may be necessary to resolve questions
concerning the effects of welding exposures on fertility.

20. Teeth
Tatintsyan and Abgaryan (Ref. 197) surveyed electric
arc welders and determined that 87% have some degree
of periodontal disease and gingivitis. To determine
whether the problem was related to a lack of defenses
against bacterial growth in the mouth, they measured the
concentration of lysozyme in saliva collected from each
of 35 welders and 28 nonwelding controls. Lysozyme, an
enzyme destructive to the cell wall of some bacteria, is a
normal constituent of saliva and other body fluids. It was
used as a measure of the ability of the body to repress the
population of micrococcus lysodenticus in the mouth.
Average lysozyme concentrations were 167.5 8 g/ml
in the control population and 103.7 13 g/ml in welders. These investigators developed a small filtration device to protect the mouth from exposure to welding
fumes. After using the device for 2 years, lysozyme increased from 103.7 to 148.5 g/ml in welders. In addition, use of this device reduced the concentrations of Mn
and Zn in saliva.

21. Effects of Specific Metals

21.1 Beryllium. Acute berylliosis, caused by brief exposure to high concentrations of beryllium fumes, is characterized by inflammation of the nasopharynx, trachea,
and bronchi, with possible development of pulmonary
edema. Chronic berylliosis, caused by long-term exposure to low concentrations of beryllium is typified by a
granulomatous process in the lung which may not become apparent until months or years after exposure
ceases. Either can be fatal.
A typical case of chronic beryllium disease in a fiftyone-year-old Scottish welder was described by Monie
and Roberts (Ref. 138). The welder had a long-standing
cough with morning production of mucoid phlegm; his
chest x-ray showed reticulonodular shadowing. Examination showed reduced pulmonary function, and biopsies
revealed mild fibrosis and epithelioid cell granulomas.
The identification of beryllium in biopsied lung tissue,
led to the diagnosis of this condition as chronic beryllium
disease producing a granulomatous pneumonitis. The
electrical firm where he had worked had closed, and it

50

could not be confirmed that he had worked with beryllium. However, the Heath and Safety Executive of Scotland confirmed that the Sciacky process, with which he
had worked, used a consumable electrode of a copper
alloy containing between 2 and 3% beryllium.
21.2 Cadmium. Acute inhalation of cadmium fumes can
cause metal fume fever-like symptoms with fever, headache, shortness of breath, chest pain, abdominal pain,
vomiting, diarrhea, irritation of the mucosa, and muscle
pain. In addition, cadmium is a severe respiratory irritant
and exposure can cause pulmonary edema (accumulation
of fluids in the lung) within hours after exposure, followed by chemical pneumonitis.
Yates and Goldman (Ref. 225) described a case of
acute cadmium poisoning who had been brazing ship
propellers with an oxyacetylene torch using an alloy containing 20% cadmium propellers in a confined space
with poor ventilation. Ten days after working in these
conditions, he began to experience malaise and breathlessness and developed fever and joint pain. A physical
examination revealed widespread inspiratory crackles
over both lower lobes, and a chest X-ray showed mottled
shadowing in the lungs. Lung function was restricted
(FEV1 and FVC were reduced). With prednisone therapy,
his lesions resolved completely within several months.
Fuortes et al. (Ref. 67) described an acute fatality in a
man who was using a propane torch and a soldering gun
to join sheet metal surfaces in an unventilated garage.
After three days of this activity, he developed an extremely high fever, a cough, and abdominal pain. He died
three days later in the hospital. Autopsy revealed pulmonary edema and congestion. Exceedingly high cadmium
levels (280 ng/ml) were discovered during a heavy metal
screen of cardiac blood (allowable occupational blood
levels are <5 ng/ml). Milligram quantities of cadmium
were identified by AAS on the soldering gun tip and
sheet metal samples with which he had been working.
21.3 Iron. Noting that iron constitutes up to 50% of the
solid component of welding fumes, Lubianova (Ref.
126) conducted a survey of iron content and transferrin,
an iron binding and transport protein, in the blood of
welders. Welders were divided into groups as follows: 1)
healthy high-alloy steel welders; 2) healthy medium- and
low-alloy steel welders; 3) healthy cast iron welders; 4)
welders with pneumoconiosis; and 5) welders with
chronic bronchitis. Almost half the welders had elevated
blood iron levels when compared with nonwelding controls. Welders without pneumoconiosis had less blood
iron than welders with pneumoconiosis. The serum of
welders had twice the iron-binding capacity of nonwelding controls, due to the increase of transferrin. In all but
welders with pneumoconioses, the iron saturation of
transferrin was less than controls. The investigators con-

cluded that in welders, the overburden of iron is accommodated by increases in transferrin levels.

22. Biological Monitoring

The quantity of airborne substances to which workers
are exposed is determined by measuring breathing zone
concentrations of pollutants. Measurement of chemicals
in body fluids (e.g., blood, urine, expired air) allows the
determination of the amount of a compound that is actually absorbed by the body. Both of these techniques are
important for monitoring and controlling occupational
and environmental exposures. A third technique, identification of biomarkers, is useful in epidemiologic studies
for demonstrating that exposure to specific chemicals has
occurred. Biomarkers may detect low levels of exposure
but are generally less quantitative than data obtained by
biological monitoring. Typical biomarkers are hair concentrations of metals and chemical modifications of
DNA or protein in blood or tissues. Both biological monitoring and biomarkers are considered below.
22.1 Chromium: Biomarkers. The toxicity of chromium is related to its valence state. Hexavalent chromium is more toxic than trivalent chromium and some
hexavalent, but not trivalent, chromium compounds are
mutagenic and carcinogenic. Hexavalent chromium enters cells much more readily than trivalent chromium,
which may account for differences in their toxicity. Once
in the cell, hexavalent chromium is reduced to the trivalent form which, along with intermediate oxidation
states, is thought to be ultimately responsible for chromium carcinogenicity.
Chromium compounds induce DNA lesions which are
thought to be related to their mutagenicity and carcinogenicity. In vitro and in vivo laboratory studies have generally shown that hexavalent chromium, but not trivalent
chromium, causes an increase in the frequency of sister
chromatid exchanges (SCE). An increased frequency of
SCE has also been observed in studies of welding fumes
in laboratory animals, and is generally attributed to the
hexavalent fraction. However, SCE tests on workers exposed to nickel and chromium have yielded conflicting
results.
Using SCE and alkaline filter elution, a test which detects DNA strand breakage, Popp et al. Ref. 165) attempted to resolve this conflict. Urine samples and blood
lymphocytes were obtained from 39 electric arc welders
and 18 controls. The welders were employed full time,
welding low-to medium-alloy steel with electrodes composed of 18% chromium, 8% nickel, and 6% manganese.
The average SCE frequency was significantly lower for
welders than for the controls. However, with alkaline filter elution, the elution rate through two filter types was

51

lower for DNA from welders than from controls, indicating the presence of DNA-protein cross-links. The frequency of SCE and the extent of DNA-protein crosslinking significantly correlated with urine chromium, but
not with urine nickel concentrations. DNA-protein crosslinks can reduce the frequency of SCE and the measurable frequency of strand breakage. Thus, performing
SCE analysis alone could lead to the false impression
that DNA integrity remains intact in chromium-exposed
workers. According to the authors, these results confirmed other studies that indicated an important role for
DNA-protein cross-links in the mechanism of chromium
carcinogenicity.
Coogan et al. (Ref. 50) investigated whether the chromium content of lymphocytes can serve as a biomarker
for long-term chromium exposure. The kinetics of chromium uptake in red blood cells has been well studied, but
little is known about uptake by lymphocytes. Reasoning
that, unlike red blood cells, lymphocytes are long-lived,
Coogan maintained that lymphocyte chromium may be a
better biomarker for long-term exposure than red blood
cell chromium. Isolated rat and human white and red
blood cells were exposed to radiolabeled hexavalent
chromium in the form of potassium chromate (K251CrO4)
for 2 hours. White blood cells accumulated significantly
more chromium than did red blood cells. In other experiments, chromium levels in rat red and white blood cells
were determined at 1 hour, 24 hours, and 7 days after
oral exposure or intravenous injection. White blood cells
consistently accumulated more chromium than did red
blood cells. Chromium was undetectable in white blood
cells, and present in only low levels in red blood cells,
after administration of trivalent chromium (chromic
chloride). The chromium content of red blood cells was
independent of the valence state of the administered
chromium. The investigators concluded that the exclusive accumulation of hexavalent chromium by white
blood cells supports their use as target cells in the development of biomarkers for assessing exposure to
chromium.
22.2 Chromium: Biological Monitoring. Urinary chromium levels vary positively with air concentrations and,
thus, can be used to monitor workplace exposures to
chromium fumes (Ref. 190). The kinetics of chromium
absorption and excretion have been studied under conditions of relatively high exposures which may not be applicable to monitoring workers who experience low-level
long-term exposure to chromium. This area of study is
receiving more attention along with investigations of the
potential health effects of chronic exposure to low levels
of chromium.
Bonde and Christensen (Ref. 36) determined chromium levels in body fluids from welders working with
processes that generate fumes containing low concentra-

tions of chromate. Urine, blood, and seminal fluid were

collected from 60 welders (30 worked with GTAW of
stainless steel, the others worked with SMAW or GMAW
of mild steel) and 45 nonwelding controls selected from
six plants in Aalborg, Denmark. Ambient air concentrations of welding fume particulates and chromium were
within acceptable limits. (Environmental concentrations
of particulates, total chromium, and hexavalent chromium levels were 0.94, 0.011, and 0.003 mg/m3, respectively, among stainless steel welders and 3.1, 0.003, and
0.001 mg/m3, respectively, among mild steel welders.)
Chromium concentrations were significantly higher in
blood and urine from stainless steel and mild steel welders than from controls. However, the concentrations of
chromium in blood and urine did not change across a
workshift (i.e., there was no significant difference between chromium levels in pre-shift samples collected on
Monday morning and post-shift samples collected on
Thursday of the same week). Furthermore, urine and
blood chromium concentrations did not change after a 3week vacation break from welding. The authors found
these data to be consistent with a gradual buildup of
chromium in the body during long-term welding
exposure.
High concentrations of chromium were found in seminal fluid obtained from subgroups of welders conducting SMAW of stainless-steel. However, there was an
extremely wide variation in chromium levels, which indicated that chromium in seminal fluid may have derived, in part, from nonoccupational activities. Bonde
stated that there is a need for further determinations of
chromium levels in seminal fluids since some of his
other studies (Refs. 32 and 35) suggested a reduction in
fertility in welders. In addition, attention should also be
focused on the potential risk of delayed health effects
among welders who heretofore were not thought to be at
risk from chromium exposure.
Biological monitoring of low-level chromium exposures is also being studied by Bukowski et al. (Ref. 41).
Because studies in which effects are low and difficult to
detect are easily confounded by external factors unrelated to occupational exposures, Bukowski et al. (Ref.
41) focused at first on determining nonvocational factors
which could introduce errors into biological monitoring
data. The subjects in this ongoing study are state workers
in Hudson County, New Jersey who experienced lowlevel and intermittent exposures to soils heavily contaminated with chromium. The contamination resulted from
the use of industrial wastes containing 25% hexavalent
and trivalent chromium as fill and diking material in state
and community parks.
Chromium was determined in urine and blood samples obtained from 52 subjects employed in the community and state parks. Chromium was detected in all but

52

two urine samples. Each subject completed a questionnaire addressing potential confounding variables. Age,
educational background, physical fitness, 12-hour fasting, and dietary factors did not elevate either urinary or
blood chromium concentrations. Males had slightly
higher urinary chromium levels than did females. Beer
drinking caused a significant increase in urinary chromium. Subjects who used tobacco, or who had exercised
within 24 hours before sampling, had slightly lower
chromium levels in the urine. Neither welding nor engaging in hobby activities with possible chromium exposures was associated with increased concentrations of
chromium in the urine or blood. Persons with a history of
having worked in two or more chromium industries had
nonsignificantly higher median urinary chromium concentrations. This study is important because of its general implications for studies of exposure to low levels of
chromium.
22.3 Nickel. Angerer and Lehnert (Ref. 5) determined
nickel levels in erythrocytes (red blood cells), plasma,
and urine from 103 stainless steel welders who worked
with chromium-nickel alloyed steel (18% Cr, 10% Ni,
68% Mn) at a shipyard in Germany. Thirty-nine welders
conducted SMAW, 14 used GMAW only, and 50 used
both SMAW and GMAW. The average breathing zone
concentration of nickel was 93 g/m3. All values were
below the German Technical Guidance Concentration of
500 g/m3. GMAW welders were exposed to somewhat
higher nickel levels than SMAW welders. The control
population was composed of 123 men and women with
no occupational nickel exposure.

Nickel was not detected in erythrocytes from welders or

from controls. While nickel was undetectable in control
plasma, 82 of 103 welders had measurable plasma nickel.
(Average nickel concentration in plasma = 4.9 g/L)
Nickel concentrations in control urine ranged from <0.1 to
13.3 g/L and were below 2.2 g/L in 95% of the controls.
Nickel levels were considerably higher in the urine of
welders, with the average concentration being 18.5 g/L.
Only a weak correlation was noted between nickel levels in
plasma and urine in welders. The authors calculated that
urinary nickel levels between 30 and 50 g/L would correspond to an air exposure of 500 g/m3.
22.4 Aluminum. In a study of 23 Swedish welders published in 1988 (Ref. 187), Sjogren, Elinder et al. found
that the aluminum content of urine increased with the
duration of the welding experience. This suggested that
retained aluminum has a long biological half life. Aluminum excretion by two workers from that study was further examined in later work by this group (Ref. 59).
Because of the possibility that aluminum may be stored
in bone, aluminum concentrations in bone biopsies were
also examined in these two volunteers. Both workers had
conducted GMAW for at least 20 years prior to 1984. In
1984, worker B switched from GMAW to GTAW and
worker A became a foreman. As a result of these
changes, the aluminum exposure of these two men was
considerably reduced. Aluminum concentrations in their
urine was measured four times between 1984 and 1989
(Figure 6). Urinary aluminum excretion remained high
for the 5-year period after exposure decreased. In 1989,
the aluminum concentrations were still more than

Figure 6Aluminum Concentrations in Urine from Two Welders During the

First 4 Years After Welding of Aluminum Ceased. Elinder et al., Ref. 59

53

10 times the concentration found in persons without occupational exposure (<10 g Al/L). The aluminum concentrations in bone were 18 and 29 g/g for workers A
and B, respectively, compared to normal levels in unexposed controls of 0.6 to 5 g/g. Exposures to total fumes
were measured in 1982 and averaged 8.9 and 3.0 mg/m3
for workers A and B, respectively. No further exposure
measurements were taken after 1984, so the reduction in
exposure was not documented. The investigators concluded that aluminum accumulates in the skeleton and
tissues following long-term inhalation exposure and that
the elimination of retained aluminum is very slow, on the
order of several years.
22.5 Zinc. Chughtai et al. (Ref. 49) surveyed zinc concentrations in blood from a sample of welders and nonwelders residing in the city of Hyderabad (Sindh). Zinc
concentrations were significantly higher in serum from
electric arc and gas welders than from controls.

23. Incidental Exposures

Welders occasionally experience respiratory distress,
immune reactions, or other toxic effects from exposures
to chemicals incidental to the welding process. Exposures in this category include chemicals thought to be
generated by photochemical decomposition products of
degreasing agents and pollutants generated by pyrolysis
or thermal degradation of paints or other surface coatings
or contaminants. These exposures are not inherent to
welding and, with little exception, they can be avoided
by changes in the manufacturing process and use of
proper ventilation and safety techniques.
23.1 Degreasing Agents. Degreasing agents can present
a major hazard in welding. Severe acute respiratory distress can result from such highly toxic chemicals as
phosgene and dichloroacetyl chloride, that can arise from
UV-induced photochemical decomposition of degreasing
agents such as trichloroethylene, perchloroethylene and
1,1,1-trichloroethane. This type of hazard is greatest in
shops where GMAW or GTAW is conducted since these
processes can produce high levels of UV radiation.
While, in the past, several incidents of severe respiratory
distress were thought to have resulted from the photochemical decomposition of degreasing agents, the actual
presence of phosgene and/or dichloroacetyl chloride has
been hard to substantiate.
In 1991, Sjogren (Ref. 191) reported an incident in
which a 50-year-old, highly experienced welder performed GTAW on a piece of stainless steel that had been
degreased with trichloroethylene. The welding was done
without local exhaust ventilation to avoid disturbing the
gas shield. Almost immediately after starting to weld, he

developed dyspnea. He rested, recovered, and returned to

welding with no further symptoms until 12 hours later,
when he developed severe respiratory distress and pulmonary edema. He was treated for these symptoms but
was readmitted to the hospital 10 days later with dyspnea
and respiratory failure. His exposure to trichlorethylene
was confirmed by the presence of trichloroacetic acid (a
metabolite of trichlorethylene) in his urine. Sjogren et al.
speculated that the severe pulmonary reactions might
have been due to inhalation of the photochemical decomposition products dichloroacetyl chloride and phosgene.
However, no attempt was made to determine whether
phosgene or dichloroacetyl chloride could form under
welding conditions similar to those experienced by the
affected welder.
In a similar case involving 1,1,1-trichloroethane (methylchloroform) exposure, Selden and Sundell (Ref. 181)
extensively investigated the phosgene theory but with a
frustrating outcome, illustrating the inherent problems of
retrospective exposure assessment. In this case, a 62year-old man had been welding mild steel items covered
with a drawing oil using GMAW with an argon/carbon
dioxide gas shield. 1,1,1-Trichloroethane stabilized with
dioxane was used to degrease the pieces before welding.
After welding under these conditions for 3 days, he
developed fever and severe respiratory distress and eventually died. No phosgene was detected during a reconstruction of this incident at the worksite, even in the
presence of inordinately high levels of 1,1,1-trichloroethane (740 ppm). Phosgene could only be detected
when the concentration of airborne 1,1,1-trichloroethane
reached 1000 ppm during welding. The investigators
concluded that, while the presence of phosgene could not
be confirmed, the illness, diagnosed as toxic pulmonary
edema, was related to welding in the presence of a chlorinated solvent.
23.2 Coated or Contaminated Surfaces. Two experienced welders who developed fever, spirometric deterioration, and bronchial hyperreactivity after GMAW of
steel painted with lacquers containing chlorinated polymers were described by Sjogren et al. (Refs. 185 and 8)
Backstrom et al, 1991). The first welder, a 54-year-old
Swedish male, experienced eye and throat irritation and
shortness of breath shortly after his first experience
welding steel coated with a new paint (Beckrysol
Grundfarg-BF paint). He developed a high fever later
that evening. The same symptoms occurred on six other
occasions when he welded steel coated with this paint.
Repeated spirometric testing showed volume and flow
decrements that slowly returned to normal. The second
man, a 49-year-old Swedish welder, developed fever and
shortness of breath 5 to 6 hours after welding steel that
had been coated with a paint (Realux Grund paint) similar in composition to the Beckrysol Grundfarg-BF paint

54

used by the first welder. He felt tired for several weeks

thereafter. He experienced similar symptoms on four
other occasions when he welded steel coated with the
same paint, although the symptoms were milder and of
shorter duration. Repeated spirometric measurements
showed flow and volume decrements that slowly
normalized.
Both welders were exposed to compounds generated
by pyrolysis of paints which were epoxidized vegetable
oils hardened by hexachloroendomethylene-tetrahydrophthalic acid anhydride (HETacid anhydride). Limited
tests demonstrated that HCl, but not phosgene, is released from this paint during welding. The authors concluded that the two welders had toxic alveolitis and
obstructive lung disease induced by HCl and other chlorinated compounds. They related the symptoms to reactive airways dysfunction syndrome (RADS), which is
characterized by asthma and bronchial hyperreactivity
and develops after one or more exposures to high concentrations of low-molecular weight respiratory tract irritants. Symptoms of RADS may be aggravated by
nonspecific irritants. They speculated that HETacid anhydride may have been one of the components responsible for this syndrome.
A similar case in which prolonged respiratory symptoms resulted from welding exposures was described by
Langley (Ref. 119). The 29-year-old male patient had
worked as a welder for 10 years. On one occasion, he
was brazing a piece of galvanized steel and a brass pipe
fitting using a 15% silver alloy brazing rod. TeflonTM
tape covered the connection near where he was brazing.
He was not wearing a respirator and was brazing in an
unventilated room with the windows and doors closed.
Within twenty minutes, he developed shortness of breath,
coughing, sweating, back pain, headache and a sweet
metallic taste in the mouth. Chest X-rays showed signs
of mild pulmonary edema. Two days later, shortness of
breath continued, and wheezing was present along with
increased levels of SGOT and SGPT, serum enzymes
indicative of liver injury. Breathing and pulmonary
function improved over several months. His persistent
wheezing, exacerbated by some odors and tobacco
smoke, also gradually decreased in severity. The wheezing had apparently diminished noticeably on the last examination and the patient was lost to follow-up
thereafter. The authors diagnosed this illness as RADS
associated with welding exposures. While this syndrome
is known to occur after exposure to certain irritant chemicals, it has not been associated with welding per se, and,
in this case, it may have been due to an incidental exposure to pyrolysis of the Teflon tape or to other coatings
or contaminants on the pieces being brazed.
Lipoid pneumonia results from chronic inhalation or
aspiration of mineral oil mists and has only rarely been

associated with occupational exposures. Symptoms of

diffuse, interstitial lipoid pneumonia can range from occasional cough to severe, debilitating dyspnea, fibrosis,
and loss of pulmonary function; it is occasionally fatal
(Ref. 159). Penes et al. (Ref. 162) described a welder
who developed diffuse, interstitial lipoid pneumonia attributed to exposure to mineral oil during welding. This
45-year-old patient had been exposed to oil spray for 16
years, and described his work as welding pieces of steel
covered with cutting oil. Bronchoalveolar lavage fluid
and pleural fluid were obtained from the patient, extracted with hexane, and analyzed by GC/MS. The compounds identified in these analyses were virtually
identical to those found in extracts from cutting and
stripping oils, leading to the conclusion that exposure to
cutting oils was the probable cause of his lung disease.
23.3 Allergens. Blomqvist et al. (Ref. 128) reported a
case involving a 33-year-old male welder who developed
bronchial asthma following exposure to chloramine-T, a
strong oxidizing compound with antiviral, bactericidal,
and fungicidal properties. A few minutes after welding a
container that was used for storing chloramine-T powder,
he developed rhinitis. Six months later, he was repairing
an industrial vacuum cleaner containing chloramine-T
when he experienced severe shortness of breath and
wheezing in the chest. Skin prick tests showed a positive
reaction to chloramine-T.
Kanerva et al. (Ref. 107) reported the case of a welder
who, on four separate occasions, developed an urticarial
reaction associated with high fever and facial edema
while welding mild steel profiles filled with polyurethane. Urticaria is an itchy skin rash which develops in
sensitive persons after exposure to particular drugs or
chemicals. The symptoms resolved within 48 hours. The
welder was tested for sensitivity to chemicals such as
4,4-diphenylmethane diisocyanate and 4,4-diaminophenylmethane, which could have been emitted during the
welding of polyurethane. These tests were negative and
the chemical(s) to which he was sensitized remain(s)
unknown.

Section Three
Investigations in Animals
and Cell Cultures
25. Fertility
Ernst (Ref. 63) examined the effects of trivalent and
hexavalent chromium on the reproductive system in male
Wistar rats. The animals received daily intraperitoneal
injections of chromium chloride [Cr(III)] or sodium

55

.
chromate [Cr(VI)] for 5 consecutive days. There were no
testicular changes 7 days after the last exposure. However, after 60 days, hexavalent chromium caused a reduction in testicular weight and a dose-dependent increase in
the number of atrophic seminiferous tubules. Sixty days
after administration of the highest dose tested (4 mg/kg
body weight), almost all of the seminiferous tubules
were completely degenerated and there was a marked reduction in the epididymal sperm count (Table 3). None
of these effects were seen with trivalent chromium. The
investigators suggested that the differences in toxicity
between the two valence states was related to the more
ready absorption and cell penetration of hexavalent
chromium.

26. Fibrosis
Collagen, the most abundant protein the body, forms
insoluble fibers which provide the mechanical and supportive functions of connective tissues. At least four
unique collagens (Types I through IV) are found in the
body. They differ somewhat in amino acid composition
and are found in specific tissues. When more than one
type of collagen is found in a tissue, the proportions of
the different types are characteristic and constant. A
major characteristic of lung fibrosis is a massive increase
in interstitial collagen. Lung tissue normally contains
Types I and III collagen. In a study of the process of lung
fibrosis, Yurui and Yu (Ref. 228) investigated whether
the proportions of the two types of collagen in the lung
change during the fibrotic process. They further examined whether welding fumes, which are mildly fibrogenic, and quartz, a highly fibrogenic material, produce
the same changes in collagen during fibrosis.
An enzyme-linked immunosorbent assay (ELISA)
was used to determine the content and distribution of
Type I and Type III collagen in female Wistar rats exposed by intratracheal instillation to either quartz dust or
welding fume. (Details about the welding process used to
generate the fumes were not given). With welding fumes,
the content of Type III collagen in the lung was significantly greater 30 days after exposure but Type I collagen
did not increase until about 180 days after exposure. Because of this, the ratio of Type I/Type III collagen decreased gradually for about 180 days before it began to
return to normal levels. In contrast, instillation of quartz
caused a continuous increase in both Type I and Type III
collagen. These results suggest that different mechanisms are involved in the cellular response to highly fibrotic and mildly fibrotic dusts. The authors concluded
that the ratio of Type I to Type III collagen can be used
for evaluation of the fibrogenicity of respirable dusts.

Table 3
Effect of Short-Term Exposure to Tri- and
Hexavalent Chromium on Testicular Weight,
and Number of Spermatozoa per
Epididymis in Adult Wistar Rats

Treatment
Control
Cr(III)
1 mg/kg
Cr(III)
2 mg/kg
Cr(III)
4 mg/kg
Cr(VI)
1 mg/kg
Cr(VI)
2 mg/kg
Cr(VI)
4 mg/kg

Body
Weight
(g)

Relative
Testicular
Weighta
(g/100 g)

Spermatozoa/
Epididymis
( 106)

448.87.7

1.060.04

64011

419.411.6

1.030.04

63012

457.08.0

1.050.04

63411

429.87.3

1.040.03

6299

442.09.3b

1.710.05b

36912b

367.112.7b

1.690.05b

1315b

390.910.9b

1.580.04b

497b

aTesticular

weight (g)/body weight (g)

different than control animals (p<0.01)
Data from Ernst, Ref. 63
bStatistically

Hicks and Olufsen (Ref. 89) investigated whether myofibroblasts proliferate in fibrotic processes in lungs exposed to welding fumes. Rats were exposed by
intratracheal instillation to crystalline silica or to particles from fumes generated by GTAW of mild steel. The
animals were sacrificed, and strips of lungs were removed and placed in tissue culture medium. Contractility, indicative of myofibroblast activity, could be
stimulated in fibrotic tissues from rats treated with either
silica or welding fumes. Histological examination confirmed the presence of myofibroblasts in fibrotic lung
tissue.

27. In Vitro Tests

Ozone, a strong respiratory tract irritant, is produced
in substantial amounts during hyperbaric inert gas welding. Ozone is known to reduce the phagocytic activity of
alveolar macrophages. In continuing studies of the effects of hyperbaric pressure on physiological functions,
Jakobsen et al. (Ref. 101) found that the phagocytic activity of rat alveolar macrophages is not affected by elevated pressure. However, the combined effects of
pressure and ozone caused a greater reduction in phagocytic activity than did ozone alone, indicating that in-

56

creasing pressure may increase the effects of ozone on

alveolar macrophages.
Polymorphonuclear leukocytes migrate into the lungs
and other tissues in response to the presence of foreign
materials. They play a major role in the inflammatory
process and are important in defending the body against
bacteria. When PMNs interact with fibrogenic materials,
they release superoxide which is converted by the enzyme superoxide dismutase to hydrogen peroxide. The
PMN enzyme myeloperoxidase can further convert the
hydrogen peroxide to hypochlorous acid (HOCl). These
strong oxidants contribute to the destruction of invading
bacteria, but they can also damage lung tissue. The generation of oxidants by PMNs can be assayed with the
chemiluminescent compound luminol. Upon reaction
with oxidants, luminol generates a flash of light which is
readily detected with a liquid scintillation counter. This
reaction forms the basis for the chemiluminescence assay, a widely used immunological test.
Saburova et al. (Ref. 177) used the chemiluminescence assay to determine whether welding fumes and
other fibrogenic dusts elicit the formation of HOCl by
isolated PMNs. While human PMNs generated HOCl in
response to opsonized zymosan (a complex carbohydrate
obtained from the walls of yeast cells), none of the three
fibrogenic dusts tested (quartz, natural zeolite, and welding fumes) stimulated HOCl formation. The authors concluded that HOCl does not play a role in the pathogenic
processes associated with inhaled welding fumes.
Adamis et al. (Ref. 1) compared the biological activity of fumes collected from GTAW of an aluminum-magnesium alloy with dusts collected from other operations
in an aluminum plant in Hungary. Red blood cell hemolysis, interference with macrophage metabolism, as indicated by a decrease in the reduction of the dye
triphenyltetrasolium chloride, and lysis of macrophages,
as indicated by the release of the enzyme lactic dehydrogenase into the culture medium, were determined. None
of the three samples was classified as hazardous on the
basis of the experimental results. However, welding
fume particulates were more toxic than the other dust
samples collected.

28. Effects of Welding Fumes in

Animals
Naslund et al. (Ref. 143) examined pulmonary effects
in sheep of welding fume produced by SMAW of black
iron. In the first experiment, a single dose of 0.5 gram
SMAW fume particulate was administered to sheep by
intratracheal instillation. Acute exposure to welding
fume significantly increased the mean pulmonary arterial
pressure and the number of alveolar leukocytes. Hemat-

ocrit values, arterial oxygen tension and pO2 were reduced. The sheep were killed 4 hours after exposure and
the lungs removed and analyzed for metal content. Fe,
Mg, Ti, Al, and Mn accumulated in the lungs. Mn levels
were increased 40-fold over control values.
In a second study, designed in part to examine the distribution of welding fume in the body, sheep were tracheotomized and exposed by inhalation to welding fumes 3
hours a day, 5 days a week for 25 to 33 weeks. Iron oxide
accumulation was monitored by magnetopneumography.
Sheep were killed after the last exposure and the lungs
were removed and examined for histopathological
changes. Fibrosing pneumonitis and slight emphysema
were observed in lung tissue. The concentrations of Cu,
Ti, Mg and Zn were determined in skeleton, kidney, lung
and liver. None of the metals was elevated in bone or
liver. The concentrations of Mg and Mn were slightly increased in the kidney. Substantial amounts of Mn and
iron accumulated in the lungs. None of the metal concentrations in the blood or lymph was significantly altered
by exposure.
Pokrovskaia and Cherednichenko (Ref. 163) compared the effects of fumes from five different welding
electrodes on the cardiovascular system and respiratory
tract in rats. Fumes (12.5 mg/kg per dose) were administered by intratracheal instillation once weekly for 4
weeks. The proportions of five elements among the different fumes were: Mn (7.5 to 11%), Si (1.9 to 18.9%), K
(3.5 to 27.1%), Na (5.0 to 15.3%), Ca (1.4 to 22.0%), F
(9.6 to 20.0%), and Fe (18.0 to 29.9%); details on the
composition of the individual welding fumes were not
given. Rats were sacrificed at 1, 3 and 6 months after
treatment, and tissues were examined by histopathological techniques. After 1 month, swelling of the bronchial
epithelium, atelectasis (poor expansion of the alveoli),
and thickening of alveolar walls were seen in the lungs.
Many particles were engulfed by alveolar macrophages.
Damaged muscle fibers were seen in the heart. At 3
months, there were still substantial effects in the lung and
cardiovascular system. By 6 months, much of the tissue
damage had resolved. Fibrosis was not observed at any
time. The investigators concluded that toxicity was related to the content of K, F, and soluble Mn compounds
in the fumes.

29. Biochemical Studies

In 1987, Geleskul et al. (Ref. 70) described an animal
model for testing particle-induced lipid peroxidation.
Using this model, they demonstrated that lipid peroxidation can be used as a measure of the toxicity of welding
fumes from covered electrodes. More recently, these investigators published a series of papers in which the tox-

57

icity of fumes from three different electrodes was

measured by lipid peroxidation in the liver and lung
(Refs. 71, 72, and 73). The concentrations of Mn, Si and
Fe in fumes generated by SMAW with the three electrodes are shown in Table 4. Fume samples (50 mg each)
from three different electrodes were administered to rats
by intratracheal instillation, and malondialdehyde was
determined in lung tissue or liver mitochondria. Fumes
from all three electrodes stimulated the formation of peroxides. Peroxidation was greatest in lung tissue at 1 day
after treatment and in liver tissue at 7 days after treatment. Fumes from electrode 3 caused more peroxidation
in the lungs than in the liver, while fumes from electrode
1 were more active in the liver, and those from electrode
2 were equally active in both organs. The authors deduced from the composition of the welding fumes (Table
5) that manganese is responsible for the liver toxicity and
silicon is responsible for lung toxicity. They also found
that fumes from electrode 2 caused more lipid peroxidation in the lung than did those from electrode 3 and related this to the greater solubility of fumes from
electrode 2. These simplified interpretations of the data
do not account for interactions between fume components, or the effects of trace metals or nonmetallic components in the welding fumes.
In a related report, Geleskul et al. (Ref. 74), measured
the tendency for welding fumes to cause lipid oxidation
in liver mitochondria by determining the effects on the
glutathione/glutathione reductase antioxidant system.
Fumes generated by welding with two different electrodes were studied. Fumes from the first electrode were
high in Ca, K, and F while those from the second electrode had a high concentration of Mn and Na. Welding
fumes were administered to rats by intratracheal instillation (50 mg per rat), and the levels of glutathione peroxidase and glutathione reductase (enzymes which regulate
the levels of the antioxidant glutathione) were measured
in liver mitochondria at 1 and 4 months after exposure.
Fumes from both electrodes increased the levels of glutathione peroxidase and glutathione reductase and decreased the content of sulfhydryl groups and glutathione.
These effects were strongest with fumes from electrode
1, which the investigators contributed to their greater
solubility.
Velichkovskii et al. (Ref. 214) compared the toxicity
of fumes from four types of electrodes in three different
assays. The first assay measured the effects of the fumes
on the chemiluminescence assay with peritoneal macrophages. The second assay examined the potential for
producing hemolysis of red blood cells. The third assay
measured the potential of welding fumes to cause peroxi-

Table 4
Metal Content of Fumes from Electrodes
Tested in Lipid Peroxidation Assays
Metal Content in Fume

Relative Toxicity

Fume
Sample

Mn

Si

Fe

Lung

Liver

16.3

2.0

44.4

18.3

17.4

24.5

3.6

17.5

48.8

Data from Geleskul et al., Refs. 71, 72, and 73

dation of lipids in the lung. For this assay, welding fumes

were administered to animals by intratracheal instillation, and lungs were collected at 1, 7, and 15 days after
treatment. Lipid peroxidation was measured by determining the content of malondialdehyde in the lungs.
Fumes from electrodes numbered 2 and 4 were substantially more toxic than those from the other two electrodes in all three assays. The investigators concluded
that there is a strong correlation between the toxic activity of the welding fumes in in vitro assays and in tests
performed in vivo. In addition, because the silicon content was highest in the fumes from electrodes 2 and 4,
while the manganese and iron content did not correlate
with toxicity, they concluded that the toxicity was related
to the silicon. As discussed above, these assumptions are
greatly simplified, in part because they ignore the effects
of trace elements. Only the concentrations of five elements in the welding fumes were determined, and it is
possible that trace elements were responsible for the
toxicity.
Using the lipid peroxide techniques described by
Geleskul et al., (Ref. 70), Kuchuk et al. (Ref. 117) measured malondialdehyde formation in liver tissue from
rats treated by intratracheal instillation with fumes from
six different electrodes. Chemiluminescence in white
blood cells isolated from these animals was also measured. The experimental conditions and the chemical
compositions of the fumes were the same as those described in the study by Pokrovskaia and Cherednichenko
(1990). Kuchuk et al. concluded that the biological activity is determined primarily by the Mn and Fe content but
that it also increases with the concentration of other ingredients as follows (in decreasing order): K, Si, F,
Na/Ca. These conclusions are contrary to those of Velichkovskii, (1990) who found no correlation between toxicity and Mn or Fe content of welding fumes.

References

(12) Balchin, N.C. Health and safety in the fabrication industry. Weld. Met. Fabr. 59: 192196, 1991.
(13) Banerjee, A. Effectiveness of eye protection in
the metal working industry. Br. Med. J. 301(6753): 645
646, 1990.
(14) Barnes, G.C. Control of fumes in the welding
environment. Svetsaren ISS 1: 1618, 1990.
(15) Barth, C., Knuschke, P., and Barth, J. Ultraviolet exposure in the environment of welding work sites.
Z. Gesamte Hyg. Grenzgeb. 36(12): 654655, 1990.
(16) Bastuji-Garin, S., Richardson, S., and Zittoun,
R. Acute leukaemia in workers exposed to electromagnetic fields. Eur. J. Cancer 26(11): 11191120, 1990.
(17) Battistoni, P., Fava, G., Arcaleni, R., and Ceccarelli, G. Evaluation of fume level emissions of an automatic welding system. Int. J. Environ. Stud. 36(4):
279284, 1990.
(18) Bauer, H.D., Dahmann, D., Fricke, H.H., Herwald, S., and Neidhart, B. Quality control of measuring
procedures by interlaboratory tests. Chromium (VI)-species in welding fumes. Staub-Reinhalt. Luft. 51(4): 125
128, 1991.
(19) Bauer, L. Welding fumes spark controversy.
Weld. Des. Fabr. 63(8): 2426, 1990.
(20) Bauer, L. Worker safety: the pressing issues.
Weld. Des. Fabr. 63(11): 4344, 1990.
(21) Bauer, L. How to select exhaust equipment.
Weld. Des. Fabr. 64(7): 2731, 1991.
(22) Beal, M.F., Mazurek, M.F., Ellison, D.W.,
Kowal, N.W., Solomon, P.R., and Pendlebury, W. Neurochemical characteristics of aluminum-induced neurofibrillary degeneration in rabbits. Neurosci 29(2):
339346, 1989.
(23) Beaumont, J.J. and Weiss, N.S. Lung cancer
among welders. J. Occup. Med. 23: 839844, 1981.
(24) Becker, N., Claude, J., and Frentzel-Beyme, R.
Cancer risk of arc welders to fumes containing chromium and nickel. Scand. J. Work Environ. Health.
11(2): 7582, 1985.
(25) Becker, N., Chang-Claude, J., and FrentzelBeyme, R. Risk of cancer for arc welders in the Federal

(1) Adamis, Z., Kurthy Komlosi, J., and Tassy, M.M.

Investigation of settling and airborne dusts collected in
plants of aluminum industry aiming at environment and
labor protection, Part 2. Magy. Alum. 27(1112): 400-6,
1990.
(2) Ahrens, W., Jockel, K.H., Patzak, W., and Elsner,
G. Alcohol, smoking, and occupational factors in cancer
of the larynx: a case-control study. Am. J. Ind. Med.
20(4): 47793, 1991.
(3) Alekseeva, I.S., Belov, V.N., and Kolodin, E. A.
Personal protection from arcing and aerosol during
welding. Svar. Proizvod. ISS 12: 2628, 1991.
(4) American Welding Society. Fumes and gases in
the welding environment, 161 Miami: American Welding Society, 1979.
(5) Angerer, J., and Lehnert, G. Occupational
chronic exposure to metals: II. Nickel exposure of stainless steel welders: biological monitoring. Int. Arch. Occup. Environ. Health 62(1): 710, 1990.
(6) Anonymous. Take precautions when cutting
PVC-Coated steel. Weld. J. 70(2): 5556, 1991.
(7) Armon, C., Kurland, L.T., Daube, J.R., and
OBrien, P.C. Epidemiologic correlates of sporadic
amyotrophic lateral sclerosis. Neurology 41(7): 1077
1084, 1991.
(8) Backstrom, I., Fryk, G., Jakobsson, R., Milerad,
E., Plato, N., Sjogren, B., and Tornling, G. Impaired
pulmonary function after welding of painted steel.
Lakartidningen 87(44): 36223623, 1990.
(9) Bakholdt, V.T. and Illum, D. Occupational eye
injuries. Ugeskr. Laeger 152(13): 917919, 1990.
(10) Bakke, S., Baste, V., Hanoa, R., and Gulsvik, A.
Prevalence of obstructive lung disease in a general population: relation to occupational title and exposure to
some airborne agents. Thorax 46(12): 863870, 1991.
(11) Bakke, O., Gulsvik, A., Lileng, P., Overa, O.,
Hanoa, R., and Eide, G.E. A postal survey on airborne
occupational exposure and respiratory disorders in Norway: causes of and consequences of nonresponse. J. Epidemiol. Community Health 44: 316320, 1990.
59

60

Republic of Germany: results of a second follow up

(19838). Br. J. Ind. Med. 48(10): 67583, 1991.
(26) Benhamou, S., Benhamou, E., and Flamant, R.
Occupational risk factors of lung cancer in a French
case-control study. Br. J. Ind. Med. 45: 231233, 1988.
(27) Blanc, P., Wong, H., Bernstein, M.S., and
Boushey, H.A. An experimental human model of metal
fume fever. Ann. Intern. Med. 114(11): 930936, 1991.
(28) Blomqvist, A.M., Axelsson, I.G.K., Danielsson,
D., Kiviloog, J., Ulander, A., and Zetterstrom, O.
Atopic allergy to chloramine-T and the demonstration
of specific IgE antibodies by the radioallergosorbent
test. Int. Arch. Occup. Environ. Health 63(5): 363365,
1991.
(29) Blot, W.J., Morris, L.E., Stroube, R., Tagnon, I.,
and Fraumeni, J.F., Jr. Lung and laryngeal cancers in relation to shipyard employment in coastal Virginia. Natl.
Cancer Inst. 65: 571575, 1980.
(30) Blot, W.J., Harrington, J.M, Toledo, A., Hoover,
R., Heath, C.W, Jr., and Fraumeni, J.F., Jr. Lung cancer
after employment in shipyards during World War II.
New Engl. J. Med. 299: 620624, 1978.
(31) Bogadi-Sare A. Respiratory disorders in stainless steel workers. Arh. Hig. Rada. Toksikol. 41(3):
249256, 1990.
(32) Bonde, J.P. Subfertility in relation to welding.
A case referent study among male welders. Dan. Med.
Bull. 37(1): 105108, 1990.
(33) Bonde, J.P. Semen quality and sex hormones
among mild steel and stainless steel welders. A cross
sectional study. Br. J. Ind. Med. 47(8): 508514, 1990.
(34) Bonde, J.P. Semen quality in welders before
and after three weeks of nonexposure. Br. J. Ind. Med.
47(8): 515518, 1990.
(35) Bonde, J.P., Hansen, K.S., Levine, R.J. Fertility
among Danish male welders. Scand. J. Work Environ.
Health 16(5): 315322, 1990.
(36) Bonde, J.P. and Christensen, J.M. Chromium in
biological samples from low-level exposed stainless steel
and mild steel welders. Arch. Environ. Health 46(4):
225229, 1991.
(37) Boshnakova, E. and Karev, G. Frequency and
structure of reproductive failures in the families of welders. Genet. Sel. 24(4): 283288, 1991.
(38) Brehme, H., Farwer, A., and Sroka, U. Computer-aided ozone measurements in gas-shielded welding. Gas Aktuell. 40: 2733, 1990.
(39) Breslow, L., Hoaglin, L., Rasmussen, G., and
Abrams, H.K. Occupations and cigarette smoking as
factors in lung cancer. Am. J. Public Health 44: 171
181, 1954.
(40) Buiatti, E., Kriebel, D., Geddes, M., Santucci,
M., and Pucci, N. A case control study of lung cancer in

Florence, Italy. I. Occupational risk factors. J. Epidemiol. Community Health 39: 244250, 1985.
(41) Bukowski, J.A., Goldstein, M.D., Korn, L.R.,
and Johnson, B.B., Biological markers in chromium exposure assessment: confounding variables. Arch. Environ. Health 46(4): 230236, 1991.
(42) Bunin, G.R., Nass, C.C., Kramer, S., and Meadows, A.T. Parental occupation and Wilms tumor: results of a case-control study. Cancer Res. 49: 725729,
1989.
(43) Bunin, G.R., Petrakova, A., Meadows, A.T.,
Emanuel, B.S., Buckley, J.D., Woods, W.G., and Hammond, G.D. Occupations of parents of children with retinoblastoma: a report from the Childrens Cancer Study
Group. Cancer Res. 50(22): 71297133, 1990.
(44) Bussenius, S., Delineating danger zones around
vessels containing combustible liquids and gases. Schweissen Schneiden 43(9): 538541, 1991.
(45) Butenko, E.P., Taubenshlak, A.S., and Yavno,
E.I. Small-scale filtering apparatus for air purification.
Svar. Proizvod. ISS 3: 1920, 1990.
(46) Carmichael, A.B. and Haynes, M.J. Radiation
protection aspects of repairing a highly contaminated tritium cold box. Health Phys. 61(5): 665670, 1991.
(47) Chiesura, P., Trevisan, A., Gori, G.P., Buzza, A.,
and Calzavara, V. On the risk of cadmium poisoning in
brazing operations. Med. Del. Lav. 75(4): 300305,
1984.
(48) Chinn, D.J., Stevenson, I.C., and Cotes, J.E.
Longitudinal respiratory survey of shipyard workers:
effects of trade and atopic status. Brit. J. Ind. Med.
47(2): 8390, 1990.
(49) Chughtai, L.A., Soomro, A.L., Shaikh, D.M.,
Seehar, G.M., and Ansari, A.F. Detection of elevated
levels of zinc in the blood sera of welders. Pak. J. Zool.
23(2): 186188, 1991.
(50) Coogan, T.P., Squibb, K.S., Motz, J., Kinney,
P.L., and Costa, M. Distribution of chromium within
cells of the blood. Toxicol. Appl. Pharmacol. 108(1):
157166, 1991.
(51) Cornu, J.C. and Muller, J.P. MIG-MAG welding torches with fume exhaust devices exhaust efficiency measurement method study of relevant
parameters. Cahiers de Notes Documentaires Securite
et Hygiene du Travail 145: 663669, 1991.
(52) Cotes, J. Occupational health today and tomorrow: a view from two shipyards. J. R. Coll. Phys. 22(4):
232236, 1988.
(53) Demers, R.Y., Neale, A.V., Robins, T., and Herman, S.C. Asbestos-related pulmonary disease in boilermakers. Am. J. Ind. Med. 17(3): 327339, 1990.
(54) Dietz, A., Heller, W.D., and Maier, H. Epidemiologic aspects of cancers of the head-neck area. Offentl.
Gesundheitswes 53(10): 674680, 1991.

61

(55) Dryson, E.W. and Rogers, D.A. Exposure to

fumes in typical New Zealand welding operations. N.Z.
Med. J. 104(918): 365367, 1991.
(56) Dunn, J.E., Jr. and Weir, J.M. Cancer experience of several occupational
groups followed prospectively. Am. J. Public Health 55:
13671375, 1968.
(57) Efimov, A.A. Universal gas-exhaust protective
chamber for automated welding
apparatus. Svar. Proizvod. ISS 3: 20, 1990.
(58) Eichhorn, F. and Nies, H. Further development
of narrow-gap submerged arc welding with two wire
electrodes. Schweissen Schneiden 42(3): 125130,
1990.
(59) Elinder, C.G., Ahrengart, L., Lidums, V., Pettersson, E., and Sjogren, B. Evidence of aluminium accumulation in aluminium welders. Br. J. Ind. Med. 48(11):
735738, 1991.
(60) Engel, K., Hampe, A., Seebaum, D., Vinke, I.,
Wittbecker, J., Koch, R., and Trasser, F. Review of scientific research in the Federal Republic of Germany concerning emissions in laser material processing. In:
Proceedings of the International Laser Safety Conference, Cincinnati, Ohio, November 2730, 1990, 5:29
5:52. Ed. S.S. Charscan. Orlando, Florida: Laser Institute
of America, 1991.
(61) Engstrom, B., Henriks-Eckerman, M.L., and
Anas, E. Exposure to paint degradation products when
welding, flame cutting, or straightening painted steel.
Am. Ind. Hyg. Assoc. J. 51(10): 561565, 1990.
(62) Epstein, J.H., Ormsby, A., and Adams, R.M.
Occupational skin cancer. In: Occupational Skin Disease, 2nd Edition, 136159. Ed. R.M. Adams. Philadelphia: W.B. Saunders Company, 1990.
(63) Ernst, E. Testicular toxicity following shortterm exposure to trivalent and hexavalent chromium an
experimental study in the rat. Toxicol. Lett. 51: 269
276, 1990.
(64) Fisher, E.W. and Gardiner, Q. Tympanic membrane injury in welders: is prevention neglected? J. Soc.
Occup. Med. 41(2): 8688, 1991.
(65) Fletcher, A.C. and Ades, A. Lung cancer mortality in a cohort of English foundry workers. Scand. J.
Work Environ. Health 10: 716, 1984.
(66) Franchini, I. and Mutti, A. Selected toxicologic
aspects of chromium (VI) compounds. Sci. Total Environ. 71: 379387, 1988.
(67) Fuortes, L., Leo, A., Ellerbeck, P.G., and Friell,
L.A. Acute respiratory fatality associated with exposure
to sheet metal and cadmium fumes. J. Toxicol. Clin.
Toxicol. 29(2): 279288, 1991.
(68) Gallagher, R.P., Salisbury, D., Band, P.R., and
Threlfall, W.J. Lung cancer in metal workers. J. Occup. Med. 33(8): 901902, 1991.

(69) Gallagher, R.P., McBride, M.L., Band, P.R.,

Spinelli, J.J., Threlfall, W.J., and Tamaro, S. Brain cancer and exposure to electromagnetic fields. J. Occup.
Med. 33(9): 944945, 1991.
(70) Geleskul, Y.F., Gorban, L.N., Baida, N.A., Voitkevich, V.G., and Kornuta, N.A. Peroxidation of lipids
in animal tissues as an index of biological aggressiveness
of welding aerosols. Vrach. Delo. 6: 108110, 1987.
(71) Geleskul, Y.F., Kornuta, N.A., Voitkevich, V.G.,
Borisenko, S.N., and Troitskii, G.V. Effect of metals in
silicon-containing welding dust on the state of lipid peroxidation in white rat tissue. Dokl. Akad. Nauk. Ukr.
SSR. Ser. B. Geol. Khim. Biol. Nauki. ISS 7: 7477,
1990.
(72) Geleskul, Y.F., Kornuta, N.A., Voitkevich, V.G.,
Borisenko, S.N., and Troitskii, G.V. Effect of welding
dust manganese on the state of lipid peroxidation in
animal tissues. Dokl. Akad. Nauk. Ukr. SSR. Ser. B.
Geol. Khim. Biol. Nauki. ISS 11: 6568, 1990.
(73) Geleskul, Y.F., Kornuta, N.A., Voitkevich, V.G.,
Borisenko, S.N., and Troitskii, G.V. The effect of silicon
from welding dust on the status of lipid peroxidation in
albino rat tissues. Dokl. Akad. Nauk. Ukr. SSR. Ser. B.
Geol. Khim. Biol. Nauki. ISS 12: 4043, 1990.
(74) Geleskul, Y.F., Volkova, N.N., Voitkevich, V.G.,
Borisenko, S.N., and Troitskii, G.V. Activity of enzymes of the glutathione antioxidant liver system in rats
as affected by welding dusts. Dokl. Akad. Nauk. Ukr.
SSR. Ser. B. Geol. Khim. Biol. Nauki. ISS 1: 6163,
1990.
(75) Gerhardsson, G. Welders health and wellbeing. Svetsaren. ISS 1: 36, 1990.
(76) Gerin, M., Siemiatycki, J., Begin, D., Kemper,
H., Lakhani, R., Nadon, L., and Richardson, L. Epidemiological monitoring system for identifying carcinogens in the Montreal work environment: a progress
report. Trav. Sante. 2: S42S46, 1986.
(77) Giroux, A.R. and Remba, M.J. Contact lens fusion to corneas by welding arcs: a hoax revisited [editorial]. J. Am. Optom. Assoc. 61(3): 162164 (1990).
(78) Gorban, L.N., Ryazanov, A.V., Voloboeva, A.A.,
Babii, V.M., and Rybalka, V.I. Comparative occupational hygiene characteristic of electrodes with a coating
based on wastes from abrasive trimming of rolled stock.
Svar. Proizvod. ISS 1: 2930, 1991.
(79) Gorban, L.N., Kucheruk, T.K., Tikhonchuk, P.P.,
and Bova, V.I. Hygienic effectiveness of removal of
welding aerosols by Kemper filter-ventilation units.
Svar. Proizvod. ISS 11: 3233, 1991.
(80) Goschnick, J., Lipp, M., Schuricht, J., Schweiker, A., and Ache, H.J. Secondary neutral mass
spectrometry as a new tool for depth resolved analysis of
particulate matter. J. Aerosol Sci. 22 (Suppl. 1): S835
S839, 1991.

62

(81) Grekula, A., Ristolainen, E., Tanninen, V.P., Hyvarinen, H.K., and Kalliomaki, P.L. Surface and bulk
chemical analysis on metal aerosols generated by manual
metal arc welding of stainless steel. J. Aerosol Sci. 17:
19, 1986.
(82) Hagberg, M., Nystrom, A., and Zetterlund, B.
Recovery from symptoms after carpal tunnel syndrome
surgery in males in relation to vibration exposure. J.
Hand Surgery. 16A(1): 6671, 1991.
(83) Hanke, C. and Karsten, H. Cataract in welders. Contribution of occupational exposure, references
for recognition as an occupational disease and presentation of preventive measures. Z. Gesamte Hyg. 36(2):
1103, 1990.
(84) Henriks-Eckerman, M.L., Engstrom, B., and
Anas, E. Thermal Degradation Products of Steel Protective Paints. Am. Ind. Hyg. Assoc. J. 51(4), 241244
(1990).
(85) Hernberg, S., Collan, Y., Degerth, R., Englund,
A., Engzell, U., Kuosma, E., Mutanen, P., Nordlinder,
H., Sand Hansen, H., Schultz-Larsen, K., Sogaard, H.,
and Westerholm, P. Nasal cancer and occupational exposures. Preliminary report of a joint Nordic case-referent study. Scand. J. Work Environ. Health 9: 208213,
1983.
(86) Hernberg., S., Westerholm, P., Schultz-Larsen,
K., Degerth, R., Kuosma, E., Englund, A., Engzell, U.,
and Hansen, H., and Mutanen, P. Nasal and sinosal cancer. Connection with occupational exposures in Denmark, Finland and Sweden. Scand. J. Work Environ.
Health 9: 315326, 1983.
(87) Hewitt, P.J. and Hirst, A.A. Development and
validation of a model to predict the metallic composition
of flux cored arc welding fumes. Ann. Occup. Hyg.
35(2): 223232, 1991.
(88) Heydon, J.L. and Kagan, A.N. Metal fume fever. N.Z. Med. J. 103(52): 883, 1990.
(89) Hicks, R. and Olufsen, A.S. Myofibroblast-mediated contracture of silicotic and metal-fume treated rat
lung strip preparations. Eur. J. Pharmacol. (Netherlands) 183(4): 15261527, 1990.
(90) Hietanen, M. and von Nandelstadh, P. Scattered
and plasma-related optical radiations associated with industrial laser processes. In: Proceedings of the International Laser Safety Conference, Cincinnati, Ohio,
November 2730, 1990, 5:29-5:52. Ed. S.S. Charscan.
Orlando, Florida: Laser Institute of America, 1991.
(91) Hilton, D.E. and Plumridge, P.N. Particulate
fume generation during GMAW and GTAW. Weld. Met.
Fabr. 59(10): 555560, 1991.
(92) Hindsen, M., Bruze, M., and Trulsson, L. Harlequin dermatitis. Contact Dermatitis 23(4): 283, 1990.
(93) Holly, E.A., Aston, D.A., Char, D.H., Kristiansen, J.J., and Ahn, D.K. Uveal melanoma in relation

to ultraviolet light exposure and host factors. Cancer

Res. 50(18): 773777, 1990.
(94) Howe, G.R., Burch, J.D., Miller, A.B., Cook,
G.M., Esteve, J., Morrison, B., Gordon, P., Chambers.,
L.W., Fodor, G, and Winsor, G.M. Tobacco use, occupation, coffee, various nutrients, and bladder cancer. J.
Natl. Cancer Inst. 64: 701713, 1983.
(95) Hu, H. Effects of ultraviolet radiation. Med.
Clin. N. Am. 74(2): 509514, 1990.
(96) Hyytiainen, K. and Saarela, K.L. Comparison
of methods to identify risk factors for low back injuries
at work. J. Occup. Accid. 11(3): 157170, 1990.
(97) Hyytiainen, K. and Uutela, A. Work factors related to stress and recurrent low back pain among Finnish planners and workers. Work Stress 5(3): 197204,
1991.
(98) IARC, Chromium. In: IARC Monographs on
the Evaluation of Carcinogenic Risks to Humans. Chromium, Nickel and Welding, 49-256. Geneva, Switzerland:
World Health Organization, 1990.
(99) IARC, Nickel. In: IARC Monographs on the
Evaluation of Carcinogenic Risks to Humans. Chromium, Nickel and Welding, 257445. Geneva, Switzerland: World Health Organization, 1990.
(100) IARC, Welding. In: IARC Monographs on
the Evaluation of Carcinogenic Risks to Humans. Chromium, Nickel and Welding, 447525. Geneva, Switzerland: World Health Organization, 1990.
(101) Jakobsen, K., Hag, R., Lindrup, A.G., and
Kambestad, B.K. Effect of ozone on the phagocytosis of
alveolar macrophages at 3.1 MPa and 40 kPa oxygen.
Joint Meeting on Diving and Hyperbaric Medicine,
Amsterdam, Netherlands, August 1118 (1990). Undersea Biomed. Res. 17(Suppl.): 122123, 1990.
(102) Jakubcik, L. Implementation of ventilation for
fume removal in welding workplaces. Zvaranie 39(8):
249251, 1990.
(103) Jarvholm, U., Palmerud, G., Kadefors, R., and
Herberts, P. The effect of arm support on supraspinatus
muscle load during simulated assembly work and welding. Ergonomics 34(1): 5766, 1991.
(104) Jauchem, J.R. Electromagnetic fields: is there
a danger? Lancet 336(8719): 884, 1990.
(105) Jung, K. Protective clothing against the effects
of heat. J. Occup. Accid. 11(3): 221228, 1990.
(106) Juutilainen, J., Laara, E., and Pukkala, E. Incidence of leukaemia and brain tumours in Finnish workers exposed to ELF magnetic fields. Int. Arch. Occup.
Environ. Health 62(4): 289294, 1990.
(107) Kanerva, L., Estlander, T., Jolanki, R., Lahteenmaki, M.T., and Keskinen, H. Occupational urticaria from welding polyurethane. J. Am. Acad.
Dermatol. 24(5): Pt. 2, 825826, 1991.

63

(108) Kantor, A.F., McCrea Curnen, M.G., Meigs,

J.W., and Flannery, J.T. Occupations of fathers of patients with Wilms tumour. J. Epidemiol. Community
Health 33: 253256, 1979.
(109) Kestenbaum, A. and Coyle, J. Laser system
design for safety in manufacturing. In: Proceedings of
the International Laser Safety Conference, Cincinnati,
Ohio, November 2730, 1990, 5:295:52. Ed. S.S. Charscan. Orlando, Florida: Laser Institute of America, 1991.
(110) Kilburn, K.H. and Warshaw, R. Airway obstruction in asbestosis studied in shipyard workers. Proceedings of the VIIth International Pneumoconioses
Conference, Part I. Pittsburgh, Pennsylvania, August
2326, 1988. NIOSH, U.S. Department of Health and
Human Services, DHHS (NIOSH) Publication No.
90-108 Part I, 408412, 1990.
(111) Kilburn, K.H. and Warshaw, R.H. Pulmonary
functional impairment from years of arc welding. In:
Proceedings of the VIIth International Pneumoconioses
Conference, Part II. Pittsburgh, Pennsylvania, August
2326, 1988. NIOSH, U.S. Department of Health and
Human Services, DHHS (NIOSH) Publication No.
90-108 Part II, 12641268, 1990.
(112) Kilburn, K.H., Warshaw, R., Boylen, C.T.,
Thornton, J.C., Hopfer, S.M., Sunderman, F.W., Jr., and
Finklea, J. Cross-shift and chronic effects of stainlesssteel welding related to internal dosimetry of chromium
and nickel. Am. J. Ind. Med. 17(5): 607615, 1990.
(113) Kishimoto, T., Yamaguchi, K., and Umemori,
Y. Infection in a welder with pneumoconiosis and asbestosis. Jpn. J. Chest Dis. 50(9): 768772, 1991.
(114) Kjuus, J., Skjaerven, R, Langard, S., Lien, J.T.,
and Aamodt, T. A case-referent study of lung cancer,
occupational exposures and smoking. Scand. J. Work
Environ. Health 12: 193202, 1986.
(115) Kleiner, A.I., Makotchenko, V.M., Efremova,
V.A., Kashin, L.M., and Smolnikova, L.S. The prevalence of chronic bronchitis among workers in the dustexposed occupations of machinery building. Vrach.
Delo. ISS 2: 7678, 1991.
(116) Krishnan, S.S., McLachlan, D.R., Krishnan, B.,
Fenton, S.S.A., and Harrison, J.E. Aluminum toxicity to
the brain. Sci. Total Environ. 71: 5964, 1988.
(117) Kuchuk, A.A., Stezhka, V.A., Gorban, L.N.,
and Blakita, I.V. Modern methods for toxicological and
hygienic evaluation of new welding materials. Svar.
Proizvod. ISS 5: 278, 1991.
(118) Lam, H.F., Conner, M.W., Rogers, A.E.,
Fitzgerald, S., and Amdur, M.O. Functional and morphologic changes in the lungs of guinea pigs exposed to
freshly generated ultrafine zinc oxide. Toxicol. Appl.
Pharmacol. 78(1): 2938, 1985.

(119) Langley, R.L. Fume fever and reactive airways dysfunction syndrome in a welder. Southern. Med.
J. 84(8): 10341036, 1991.
(120) Lasfargues, G., Phan Van, J., Lavandier, M.,
Renault, B., Renjard, L., Moline, J., Rondeau du Noyer,
C., and Lemarie, E. Pulmonary siderosis and long-term
respiratory risks of arc welders. Rev. Mal. Respir. 8(3):
304306, 1991.
(121) Lee, H.S., Chia, S.E., Yap, J.C., Wang, Y.T.,
and Lee, C.S. Occupational asthma due to spot-welding. Singapore Med. J. 31(5): 506508, 1990.
(122) Lee, H.S., Phoon, W.H., Wang, Y.T., Poh, S.C.,
Cheong, T.H., Yap, J.C.H., Lee, F.Y.W., and Chee,
C.B.E. Occupational asthma in Singapore a review
of cases from 1983 to 1990. Singapore Med. J. 32: 398
402, 1991.
(123) Lerchen, M.L., Wiggins, C.L., and Samet, J.M.
Lung Cancer and Occupation in New Mexico. J. Natl.
Cancer Inst. 79: 639645, 1987.
(124) Lewis R. Metals. In: Occupational Medicine,
297326. Ed. J. LaDou, Appleton, and Lange. Norwalk,
Connecticut, 1990.
(125) Loriot, J. and Tourte, J. Hazards of Contact
Lenses Used by Workers. Int. Arch. Occup. Environ.
Health 62(2): 105108, 1990.
(126) Lubianova, I.P. Manifestations of excessive
iron accumulation in steel and cast iron welders. Gig.
Tr. Prof. Zabol. ISS 5: 1014, 1990.
(127) Lubianova, I.P., Kucheruk, T.K., Rossinskaia,
L.N., Riazanov, A.V., Timofeeva, N.T, Tsyrkunov, L.P,
Ishchenko, A.D., Kovalenko, G.D., Makarova, M.L., and
Makarenko, T.I. The effect of the working conditions on
the development of diseases in cast-iron welders. Vrach.
Delo. ISS 2: 6670, 1991.
(128) Lukak, M., Moravik, J., Ulrich, L., and Malik,
E., Functional changes in the respiratory tract of welders.
Prac. Lek. 43(3-4): 107110, 19911992.
(129) Malmqvist, K.G., Johansson, G., Bohgard, M.,
and Akselsson, K.R. Process-dependent characteristics
of welding fume particles. In: Health Hazards and Biological Effects of Welding Fumes and Gases. Eds. R.M.
Stern, A. Bernin, A.C. Fletcher, and J. Jarvisalo. Amsterdam: Exerpta Medica, 3146, 1986.
(130) Mangold, V.L., Jr. Safety and robotic welding:
mutual compatibility. Weld. J. 70(9): 7174, 1991.
(131) Manz, A.F. Safety in Welding and Cutting
details practices and procedures. Occup. Health Safety
59(5): 2731, 1990.
(132) Marciniak, R. and Badowski, R. Static
changes of the spine in workers in the automobile industry. Med. Pr. 42(1): 2330, 1991.
(133) Mastromatteo, E. Nickel. Am. Ind. Hyg.
Assoc. J. 47(10): 589601, 1986.

64

(134) McMillan, G.H.G. and Pethybridge, R.J. A

clinical, radiological and pulmonary function case-control study of 135 dockyard welders aged 45 years and
over. J. Soc. Occup. Med. 34: 323, 1984.
(135) Medack, J. and Heinze, B. The emission and
composition of welding fumes generated during the
MAG welding of metal coated [steel] sheet. Weld. Int.
4(11): 908910, 1990.
(136) Melbostad, E. and Ruud, B.F. Bronchial reactivity airways symptoms and lung function in welders
and machine workers. Am. Rev. Respir. Dis. 143(4):
Part 2, A440, 1991.
(137) Mertens, J., Bubmann, M., and Reker, U.
Welding spark injuries of the ear. Observations of personal case material. Laryngorhinootologie 70(8): 405
408, 1991.
(138) Monie, R.D. and Roberts, G.H. Chronic beryllium pneumonitis: first case accepted by U.K. register
from Scotland. Scott. Med. J. 36(6): 185186, 1991.
(139) Monteagudo, F.S.E., Cassidy, M.J.D., and
Folb, P.I. Recent developments in aluminium toxicology. Med. Toxicol. 4: 116, 1989.
(140) Mori, T., Matsuda, A., Akashi, S., Ogata, M.,
Takeoka, K., and Yoshinaka, M. The three dimensional
distribution of chromium and nickel alloy welding
fumes. Acta. Med. Okayama 45(4): 233240, 1991.
(141) Moulin, J.J., Diebold, F., and Limasset, J.C.
Cancer risk among welders, a review of the epidemiological evidence. Cahiers de Notes Documentaires Securite et Hygiene du Travail 4th Quarter(145): 679691,
1991.
(142) Narda, R., Magnavita, N., Sacco, A., Sarnari,
L., and Sani, L. Chronic conjunctivitis in arc welders: a
cross-sectional study. Med. Lav. 81(5): 399406, 1990.
(143) Naslund, P.E., Andreasson, S., Bergstrom, R.,
Smith, L., and Risber, B. Effects of exposure to welding
fume: an experimental study in sheep. Eur. Respir. J.
3(7): 800806, 1990.
(144) Neki, A.S. Fluphenazine induced welding arc
maculopathy [letter; comment]. Br. J. Ophthalmol.
76(4): 255, 1992.
(145) Newhouse, M.L., Oakes, D., and Woolley, A.J.
Mortality of welders and other craftsmen at a shipyard
in NE England. Br. J. Ind. Med. 42: 406410, 1985.
(146) NIOSH FACE Report: Welder dies after falling
22 feet from a roof at a mall construction site in Maryland, November 4, 1989. Division of Safety Research,
NIOSH, U.S. Department of Health and Human Services, Morgantown, West Virginia, Report Number
FACE-90-19 (1990).
(147) NIOSH FACE Report: Journeyman ironworker
dies following a 22.5-foot fall from a walkway in Maryland, May 26, 1991. Division of Safety Research,
NIOSH, U.S. Department of Health and Human Ser-

vices, Morgantown, West Virginia, Report Number

FACE-91-18 (1991).
(148) NIOSH FACE Report: Ironworker dies in Ohio
following a 20-foot fall through a skylight opening, January 28, 1991 Division of Safety Research, NIOSH, U.S.
Department of Health and Human Services, Morgantown, West Virginia, Report Number FACE-91-11
(1991).
(149) Norn, M. and Franck, C. Long-term changes
in the outer part of the eye in welders. Prevalence of
spheroid degeneration, pinguecula, pterygium, and corneal cicatrices. Acta Ophthalmologica 69(3): 382386,
1991.
(150) Norseth, T. Metal Carcinogenesis. In: Living
in a Chemical World. Ann. N.Y. Acad. Sci. 534: 377
386, 1988.
(151) Okuno, T. Thermal effect of infra-red radiation on the eye: a study based on a model. Ann. Occup.
Hyg. 35(1): 112, 1991.
(152) Olah, L. and Pospisilova, N. Hygienic and
health problems in welding of high-alloy materials.
Zvaranie 39(4): 97102, 1990.
(153) Oleinichenko, K.A., Muratov, V.A., Solovev,
O.O., Antonenko, Z.A., and Samotkan, M.A. Effect of
the regime of surfacing with a flux-cored strip on the
emission of harmful substances in the workplace air.
Svar. Proizvod. ISS 2: 1921, 1990.
(154) Olsen, J., Sabroe, S., and Lajer, M. Welding
and cancer of the larynx: a case-control study. Eur. J.
Cancer Clin. Oncol. 20: 639643, 1984.
(155) Olshan, A.F., Breslow, N.E., Daling, J.R., Falletta, J.M. Grufferman, S., Robison, L.L., Skerwitz, M.,
and Hammond, G.D. Wilms tumor and paternal occupation. Cancer Res. 50(11): 32123217, 1990.
(156) Olsson, R., Persson, K.A., and MacKay, L.
Gas selection for increased productivity. Weld. Met.
Fabr. 59(9): 502556, 1991.
(157) Onodera, J., Yabuta, H., Nishizono, T., Nakamura, C., and Ikezawa, Y. Characterization of Aerosols
from Dismantling Work of Experimental Nuclear Power
Reactor Decommissioning. J. Aerosol Sci. 22: Suppl. 1,
S747S750, 1991.
(158) Pachocki, K.A. and Gajewski, A.K. Exposure
to electromagnetic fields and risk of leukemia. Rocz.
Panstw. Zakl. Hig. 42(3): 217221, 1991.
(159) Palmer, W.G. Hazards associated with mineral
oil smokes. In: Proceedings of the Smokes/Obscurants
Symposium XV. 1618 April, 1991, II, 753776. Laurel,
MD: Johns Hopkins University, 1991.
(160) Panio, P. Carpal tunnel syndrome can affect
welder productivity. Weld. J. 69(4): 6970, 1990.
(161) Paul R. Lead pollution of work sites in auto
body repair shops. Z. Gesamte Hyg. 36(12): 651653,
1990.

65

(162) Penes, M.C., Vallon, J.J., Sabot, J.F., and Vallon, C. GC/MS detection of paraffins in a case of lipoid
pneumonia following occupational exposure to oil
spray. J. Anal. Toxicol. 14: 372374, 1990.
(163) Pokrovskaia, T.N. and Cherednichenko, V.M.
Effects of welding dust on
morphology of the bronchopulmonary system and heart
in experimental animals. Gig. Tr. Prof. Zabol. ISS 9:
3740, 1990.
(164) Polednak, A.P. Mortality among welders, including a group exposed to nickel oxides. Arch. Environ. Health 36: 235242, 1981.
(165) Popp, W., Vahrenholz, C., Schmieding, W.,
Krewet, E., and Norpoth, K. Investigations of the frequency of DNA strand breakage and cross-linking and of
sister chromatid exchange in the lymphocytes of electric
welders exposed to chromium and nickel-containing
fumes. Int. Arch. Occup. Environ. Health 63(2): 115
120, 1991.
(166) Power, W.J., Travers, S.P., and Mooney, D.J.
Welding arc maculopathy and fluphenazine. Br. J.
Ophthalmol. 75(7): 433435, 1991.
(167) Puntoni, R., Vercelli, M., Merlo, F., Valerio, F.,
and Santi, L. Mortality among shipyard workers in
Genoa, Italy. Ann. N.Y. Acad. Sci. 330: 353377, 1979.
(168) Rakow, P.L. Contact lenses in an industrial environment. J. Ophthalmic Nurs. Technol. 9(2): 7273,
1990.
(169) Rekus, J.F. Managing welding hazards in confined spaces. Weld. J. 69(7): 2734, 1990.
(170) Richter, K. H. On the physical strain on arc
welding operators depending on work order. Schweisstechnik (Berlin) 40(7): 309314, 1990.
(171) Rinsky, R.A., Melius, J.M., Hornung, R.W.,
Zumwalde, R.D., Waxweiler, R.J., Landrigan, P.J., Bierbaum, P.J, and Murray, W.E., Jr., Case-control study of
lung cancer in civilian employees at the Portsmouth
naval shipyard, Kittery, Maine. Am. J. Epidemiol. 127:
5564, 1988.
(172) Romanenko, V.F., Calculation of the maximum collection coefficient of the welding aerosol using
burners with annular dust-gas collectors. Svar. Proizvod. ISS 5: 2829, 1990.
(173) Romanenko, V.F. and Fridman, V.M. Design
calculation of dust-gas collectors with lateral suction for
the removal of welding aerosol. Svar. Proizvod. ISS 6:
3233, 1990.
(174) Ronco, G., Ciccone, G., Mirabelli, D., Troia, B.,
and Vineis, P. Occupation and lung cancer in two industrialized areas of northern Italy. Int. J. Cancer 41: 354
358, 1988.
(175) Roquet-Doffiny, Y., Roquet, P., and Lachapelle,
J.M. Burns, artificial actinitis and multiple epitheliomas

in a cutting-welder. Dermatologica 155: 176179,

1977.
(176) Rossignol, M., Seguin, P., and DeGuire, L.
Age modification of a healthy worker effect on pulmonary fictions from exposure to welding fumes. Am. J.
Epidemiol. 132(4): 761, 1990.
(177) Saburova, V.I., Romashevskaya, E.I., and Velichkovskii, B.T. The myeloperoxidase system of polymorphonuclear leukocytes and fibrogenous dusts. Gig.
Tr. Prof. Zabol. ISS 8: 5153, 1990.
(178) Sankila, R.J., Karjalainen, E.S., Oksanen,
H.M., Hakulinen, T.R., and Teppo, L.H.I. Relationship
between Occupation and Lung Cancer as Analyzed by
Age and Histologic Type. Cancer 65(7): 16511656,
1990.
(179) Schoenberg, J.B., Stemhagen, A., Mason, T.J.,
Patterson, J., Bill, J., and Altman, R. Occupation and
lung cancer risk among New Jersey white males. J.
Natl. Cancer Inst. 79(1): 1321, 1987.
(180) Seemann, J., Witting, P., Kollmeier, H., Muller,
K.M., and Schejbal, V. Trace Metal Analysis of Chromium and Nickel in Lung Tissue Fixed and Stored in
Formalin. Pathol. Res. Pract. 186(1): 197201, 1990.
(181) Selden, A. and Sundell, L. Chlorinated solvents, welding and pulmonary edema [letter; comment].
Chest 99(1): 263, 1991.
(182) Shaikh, T.Q. and Bhojani, F.A. Occupational
injuries and perception of hazards among road-side
welding workers. J. Pak. Med. Assoc. 41(8): 187188,
1991.
(183) Silverstein, M., Maizlish, N., Park, R., and
Mirer, F. Mortality among workers exposed to coal tar
pitch volatiles and welding emissions: an exercise in epidemiologic triage. Am. J. Public Health 75: 12831287,
1985.
(184) Simonato, L., Fletcher, A.C., Andersen, A.,
Anderson, K., Becker, N., Chang-Claude, J., Ferro, G.,
Gerin, M., Gray, C.N., Hansen, K.S., et al. A historical
prospective study of European stainless steel, mild steel,
and shipyard welders. Br. J. Ind. Med. 48(3): 14554,
1991.
(185) Sjogren, B., Carstensen, J. Cancer morbidity
among Swedish welders and gas cutters. In: Health
Hazards and Biological Effects of Welding Fumes and
Gases. Proceedings International Conference, Copenhagen, 1825 Feb. 1985, 461463. Eds. R.M. Stern, A.
Berlin, and J. Jarvisalo. Amsterdam: Elsevier Science
Publishers, 1986.
(186) Sjogren, B., Hedstrom, L., and Ulfvarson, U.
Urine chromium as an estimator or air exposure to
stainless steel welding fumes. Int. Arch. Occup. Environ. Health 51: 347354, 1983.
(187) Sjogren, B., Elinder, C.G., Lidums, V. and
Chang, C. Uptake and urinary excretion of aluminum

66

among welders. Int. Arch. Occup. Environ. Health 60:

7779, (988.
(188) Sjogren, B., Gustavsson, A., and Hedstrom, L.
Mortality in two cohorts of welders exposed to high and
low-levels of hexavalent chromium. Scand. J. Work Environ. Health 13: 247251, 1987.
(189) Sjogren, B., Gustavsson, P., and Hogstedt, C.
Neuropsychiatric symptoms among
welders exposed to neurotoxic metal. Br. J. Indust. Med.
47(10): 704707, 1990.
(190) Sjogren, B., Backstrom, I., Fryk, G., Jakobsson,
R., Milerad, E., Plato, N., and Tornling, G. Fever and
respiratory symptoms after welding on painted steel.
Scand. J. Work Environ. Health 17(6): 441443, 1991.
(191) Sjogren, B., Plato, N., Alexandersson, R., Eklund, A., and Falkenberg, C. Pulmonary reactions
caused by welding-induced decomposed trichloroethylene. Chest. 99(1): 237238, 1991.
(192) Steenland, K., Beaumont, J., and Elliot, L.
Lung cancer in mild steel welders. Am. J. Epidemiol.
133(3): 220229, 1991.
(193) Steinegger, A. and Walti, K. Protection from
the ozone generated during gas shielded arc welding.
Technica 40(21): 8995, 1991.
(194) Sutton, B.R. Application of [UK] COSHH
[control of substances hazardous to health] regulations
1989 to arc welding. Stainless Today, Supplement to
Weld. Met. Fabr. 58(1): 1214, 1990.
(195) Suzuki, H., Masatoshi, T., Nakamura, Y., Kunitoshi, T., Hariguchi, S., and Nishimura, T. Activities of
lysosomal enzymes in rabbit brain with experimental
neurofibrillary changes. Neurosci. Lett. 89: 234239,
1988.
(196) Tanner, W. and Langston, J. Do environmental
toxins cause Parkinsons disease? A critical review.
Neurology 40(Suppl. 3): 1730, 1990.
(197) Tatintsyan, V.G. and Abgaryan, R.M. The condition of the teeth and periodontium in electric welders
and preventive measures in cases of their pathology.
Eksp. Klin. Med. 31(2): 166170, 1991.
(198) Ten Kate, T. and Collins, M.J. A survey of
symptoms and eye safety practices
among welders. Clin. Exp. Optom. (Australia) 73(3):
7985, 19904.
(199) Tola, S., Kalliomaki, P.L., Pukkala, E., Asp, S.,
and Korkala, M.L. Incidence of cancer among welders,
platers, machinists, and pipe fitters in shipyards and machine shops. Br. J. Ind. Med. 45: 209218, 1988.
(200) Torner, M., Zetterberg, C., Anden, U., Hansson, T., and Lindell, V. Workload and musculoskeletal
problems: a comparison between welders and office
clerks (with reference also to fishermen). Ergonomics
34(9): 11791196, 1991.

(201) Tornqvist, S., Knave, B., Ahlbom, A., and Persson, T. Incidence of leukaemia and brain tumours in
some electrical occupations. Br. J. Ind. Med. 48(9):
597603, 1991.
(202) Tossavainen, A. Estimated risk of lung cancer
attributable to occupational exposures in iron and steel
foundries. In: Complex Mixtures and Cancer Risk. Eds.
H. Vainio, M. Sorsa, and A.J. McMichael, IARC Scientific Publications, No. 104, 363-367. Lyon, International
Agency for Research on Cancer, 1990.
(203) Trent, R.B. and Wyant, W.D. Fatal hand tool
injuries in construction. J. Occup. Med. 32(8): 711714,
1990.
(204) Trevisan, A. and Maso, S. Biological indicators of the renal involvement in workers with previous
exposure to cadmium: a 5-year follow up. Med. Lav.
82(6): 542546, 1991.
(205) Tum Suden, K.D., Flynn, M.R., and Goodman,
R. Computer simulation in the
design of local exhaust hoods for shielded metal arc
welding. Am. Ind. Hyg. Assoc. J.
51(3): 11526, 1990.
(206) Ukkonen, A., Keskinen, J., and Lehtimaki, M.
Combined electrical and optical detection in continuous
mass monitoring. J. Aerosol Sci. 22(Suppl. 1): S367
S370, 1991. I.
(207) Ulrich, L., Moravik, J., Miskovic, P., Klement,
C., Malik, E., and Vladar, M. The relation between
metal exposure and immunological changes in welders.
Prac. Lek., 42(4): 147151, 1990.
(208) United States EPA. Health assessment document of nickel. Research Triangle Park, NC: U.S. Environmental Protection Agency, 1986.
(209) Valente, T. and Chiapperini, D. Evaluation of
[bodily] stress in welding by means of physiological and
thermal measurements. Riv. Ital. Saldatura 42(2): 135
138, 1990.
(210) Van der Veen, F. and Regensburg, R.E. Productivity, or quality of work as the decisive factor in
marketing ergonomics? Design considerations for a new
ergonomic welding-table. Ergonomics 33(4): 407411,
1990.
(211) Van der Wal, J.F. Exposure of Welders to
Fumes, Chromium, Nickel, Copper and Gases in Dutch
Industries. Ann. Occup. Hyg. 29(3): 377389, 1985.
(212) Van der Wal, J.F. Further studies on the exposure of welders to fumes, chromium, nickel and gases in
Dutch industries: plasma welding and cutting of stainless
steel. Ann. Occup. Hyg. 30(2): 153161, 1986.
(213) Van der Wal, J.F. Exposure of welders to
fumes and gases in Dutch industries: summary of results. Ann. Occup. Hyg. 34(1): 4554, 1990.
(214) Velichkovskii, B.T., Ibragimova, G.A.,
Korkina, L.G., Voitkevich, V.G., Geleskul, Y.F., and

67

Baida, N.A. Comparative characteristics of the biological aggressiveness of welding dusts. Gig. Tr. Prof.
Zabol. ISS 6: 1215, 1990.
(215) Verschoor, M.A., Bragt, P.C., Herber, R.F.M.,
Zielhuis, R.L., and Zwennis, W.C.M. Renal function of
chrome-plating workers and welders. Int. Arch. Occup.
Environ. Health 60: 6770, 1988.
(216) Vidmar, T.J. Select the right equipment for a
clean welding environment. Weld. J. 70(9): 4751,
1991.
(217) Wagner, U., Meerbach, W., Fischer, W., Otto,
J., Jahn, R., and Schneider, W.D. A case report of
welders lung. Z. Erkr. Atmungsorgane 174(2): 14954,
1990.
(218) Wechsler, L.S., Checkoway, H., Franklin,
G.M., and Costa, L.G. A pilot study of occupational and
environmental risk factors for Parkinsons disease. Neurotoxicology 12(3): 387392, 1991.
(219) Wedeen, R.P. and Qian, L.F. Chromium-induced kidney disease. Environ. Health Perspect. 92:
7174, 1991.
(220) Wilkins, J.R., III and Sinks, T.H., Jr. Paternal
occupation and Wilms tumour in offspring. J. Epidemiol. Community Health 38: 711, 1984.
(221) Wilson, D.F. Barrier curtains for laser hazard
control. In: Proceedings of the International Laser
Safety Conference, Cincinnati, Ohio, November 2730,
5:71-5:75. Ed. S.S. Charscan. Orlando, Florida: Laser Institute of America, 1991.
(222) Winkleman, J. Contact lenses and welding.
Arbeitsmed. Sozialmed. Praventivmed. 25(10): 456, 1990.
(223) Wisniewski, G. The effect of MAG arc welding parameters on environmental pollution. Weld. Int.
5(12): 949953, 1991.
(224) Withers, T.A., Gardiner, K., and Earwaker, L.G.
PIXE analysis of welding fumes. Nucl. Instrum. Meth.
Phys. Res., Sect. B, B56-B57, Part 2, 6736, 1991.

(225) Yates, D.H. and Goldman, K.P. Acute cadmium poisoning in a foreman plater welder. Br. J. Ind.
Med. 47(6): 429431, 1990.
(226) Young, E., van Weelden, H., van Duren, J.A.,
and Visser, A.B. Photodermatitis due to welding. Contact Dermatitis 23(4): 282, 1990.
(227) Yu, M.C., Garabrant, D.H., Hunang, T.B., and
Henderson, B.E. Occupational and other non-dietary
risk factors for nasopharyngeal carcinoma in Guangzhou, China. Int. J. Cancer 45(6): 10331039, 1990.
(228) Yurui, L. and Yu, X.H.L. A study on change
of type I and III collagen during fibrosis induced by silica and welding fume dust. Proceedings of the VIIth International Pneumoconioses Conference, Part I.
Pittsburgh, Pennsylvania, August 2326, 1988. NIOSH,
U.S. Department of Health and Human Services, DHHS
(NIOSH) Publication No. 90-108 Part I, 566570, 1990.
(229) Zaks, I.A. Sanitary-hygiene characteristics of
electrodes for arc welding steels (review). Part I. Electrodes for welding carbon and low alloy structural
steels. Svar. Proizvod. ISS 8: 3133, 1991.
(230) Zaks, I.A. Industrial hygiene characteristics of
electrodes for arc welding of steel (review). Part II. Electrodes for welding of low-alloy heat-resistant and highalloy steels. Svar. Proizvod. ISS 10: 3466, 1991.
(231) Zschiesche, W., Emmerling, G., Schaller, K.H.,
and Weltle, D. Investigations on renal impairments of
high alloy steel welders. In: Occupational and Environmental Chemical Hazards, 339343. Eds. V. Foa, E. A.
Emmett, M. Maroni, and A. Colombi Chichester. Ellis
Horwood Limited, 1987.
(232) Zyubanova, L.F., Karamyshev, V.B., Maksimenko, N.V., and Pushkin, V.Y. Hygienic evaluation of
magnetic field of suspended machines for spot welding.
Svar. Proizvod. ISS 7: 2930, 1991.